The role of HCN channels in the prenatal development of ...

The role of HCN channels in the prenatal development of the cerebral cortex in the mouse (Mus musculus,

Linnaeus 1758)

I n a u g u r a l – D i s s e r t a t i o n

zur

Erlangung des Doktorgrades

der Mathematisch-Naturwissenschaftlichen Fakultät

der Universität zu Köln

vorgelegt von

Anna Katharina Schlusche

aus Frankfurt am Main

Köln

2017

Berichterstatter: Prof.Dr.UlrichBenjaminKaupp Prof.Dr.MatsPaulssonTagdermündlichenPrüfung: 7.Dezember2016

Zusammenfassung

Zusammenfassung

Die Entwicklung des Gehirns folgt einem komplexen Muster, dessen Störung imMenschen zu einem heterogenen Spektrum an Erkrankungen, inklusive derschwerwiegendenLissenenzephalieoderMikrozephalie,führenkann.

Grundlegende Prozesse der kortikalen Entwicklung sind die Proliferation derneuralen Stamm‐ und Vorläuferzellen, deren Differenzierung in die verschiedenenneuronalen Subtypen und Gliazellen, sowie die Migration dieser differenziertenZellen zu ihren endgültigen Ziellokalisationen. Diese Vorgänge werden nebenanderen Signalen durch die elektrische Aktivität des Gehirns reguliert, welchewiederum durch die intrinsischen und extrinsischen Eigenschaften der einzelnenZellendeterminiertwird.

Ein Modulator der intrinsischen biophysikalischen Eigenschaften der Zelle ist derHyperpolarisations‐aktivierte und zyklisch Nukleotid‐modulierte (HCN) Kanal undderdurchihnvermittelteIonenstrom,Ih.

In der vorliegendenArbeitwurden zum erstenMal die Expressionsprofile der vierverschiedenen HCN‐Untereinheiten während der embryonalenVorderhirnentwicklung in der Maus analysiert. Diese Analysen offenbarten einentwicklungsspezifisches Expressionsprofil, wobei die HCN3‐ und HCN4‐SubtypenvorallemimembryonalenGehirneineRollespielen,wohingegenHCN1undHCN2diehöchsteExpressionimpostnatalenundadultenVorderhirnzeigen.

DesWeiterenkonntemittelsExpressioneinerdominant‐negativenHCNUntereinheiteine Untereinheiten‐unspezifische Ablation des durch HCN‐Kanäle vermitteltenIonenstroms (Ih) erzielt und damit die Funktion des Stroms in der embryonalenEntwicklung untersucht werden. Die embryonale Ablation des Ih führte zu einerMikrozephalie, die durch eine starke Verringerung der Proliferation und zeitlichveränderter Differenzierung der neuralen Stamm‐ und Vorläuferzellen verursachtwurde.DieMigrationderjungenNeuronehingegenwarnichtbeeinträchtigt.

In in vitro Experimenten mit primären kortikalen Stammzellkulturen konntenebenfallseinereduzierteProliferationundveränderteDifferenzierungalsFolgederpharmakologischenInhibitiondesIhnachgewiesenwerden.

Zusammenfassend zeigt diese Arbeit damit erstmalig eine entscheidende Rolle derHCN‐KanäleunddesdurchsievermitteltenIhinderProliferationundDifferenzierungvonneuralenStamm‐undVorläuferzellen.

Abstract

Abstract

The development of the brain requires a complex sequence of processes, and anydisturbances in these processes can lead to severe disorders such as lissencephalyandmicrocephaly.

The involvedevents,which include theproliferationofneural stemandprogenitorcells,thedifferentiationintoavarietyofneuronalsubtypesandglialcells,aswellasthemigration of differentiated cells to their final localization, are tightly regulated.Amongothermechanisms,theelectricalactivityofthebrain,whichisdeterminedbyextrinsicandintrinsicpropertiesofneuronsandglia,significantlycontributestothespatialandtemporalcontroloftheseprocesses.

Thehyperpolarization‐activatedandcyclicnucleotide‐gated(HCN)channelsandthecurrent(Ih)mediatedbythem,haveimportantfunctions inmodulatingthe intrinsicbiophysicalpropertiesofmanycelltypes,includingneurons.

This work presents the first detailed investigation of the endogenous subtypeexpressionofHCNsubunitsintheembryonicforebrain.Theanalysisdemonstratedasubtype switch, with HCN3 and HCN4 expression present during embryonicdevelopment, and HCN1 and HCN2 being the major subunits in the postnatal andadultmouseforebrain.

To study the functional significance of the early embryonic expression of HCNsubunits, Ihwas inhibitedbyadominant‐negativeHCN(HCN‐DN)subunit, resultingin a subunit‐unspecific ablation of the current. Forebrain‐specific embryonicexpressionof the transgenic subunitusing theEMX1promoter resulted ina severemicrocephalyphenotype.TheinvestigationofearlycorticaldevelopmentalprocessesrevealedanimpairedproliferationanddifferentiationofneuralstemandprogenitorcellsinHCN‐DN‐expressingtransgenicmice,butnoimpairedmigration.

In agreement with these findings, in vitro experiments using primary neural stemcells from the embryonic rat cortex showed a reduced proliferation and altereddifferentiationwhenIhwaspharmacologicallyblocked.

Insummary,thisisthefirstworktoshowaninvolvementoftheHCN3and/orHCN4subtype‐mediated Ih in the proliferation and differentiation of neural stem andprogenitorcellsinthemammaliancerebralcortex.

Tableofcontent

I

Tableofcontent

1 Introduction..................................................................................11.1 CorticalDevelopment................................................................................................................1

1.2 Electricalactivityindevelopmentalprocesses..............................................................4

1.2.1 Proliferation..............................................................................................................................4

1.2.2 Differentiation..........................................................................................................................6

1.2.3 Migration....................................................................................................................................7

1.3 HCNchannels................................................................................................................................9

1.3.1 StructureandfunctionofHCNchannels......................................................................9

1.3.2 ExpressionpatternsofthedifferentHCNsubtypes.............................................10

1.3.3 RegulationofHCNchannels............................................................................................11

1.3.4 KnockoutmousemodelsofthedifferentHCNsubunits.....................................13

1.3.5 HCNchannelsanddisease...............................................................................................14

1.3.6 HCNexpressionandfunctioninthedevelopingrodent.....................................14

1.4 Hypothesis...................................................................................................................................15

2 Material........................................................................................162.1 Substances...................................................................................................................................16

2.2 Buffers...........................................................................................................................................17

2.3 Antibodies...................................................................................................................................18

2.3.1 PrimaryAntibodies.............................................................................................................18

2.3.2 Secondaryantibodies.........................................................................................................19

2.4 PrimerSequences....................................................................................................................19

2.4.1 Genotyping..............................................................................................................................19

2.5 Kits..................................................................................................................................................20

3 Methods........................................................................................213.1 Animalsandhusbandry.........................................................................................................21

3.2 Genotypingoftransgenicmouselines............................................................................21

3.3 MRIAnalysis...............................................................................................................................25

3.4 QuantitativerealtimePCR...................................................................................................25

Tableofcontent

II

3.5 Nextgenerationsequencingincludingsamplepreparation.................................26

3.6 Histology......................................................................................................................................27

3.6.1 Tissuepreparation..............................................................................................................27

3.6.2 FluorescentNisslStain......................................................................................................27

3.6.3 Immunohistochemistry.....................................................................................................28

3.6.4 Microscopy.............................................................................................................................29

3.6.5 Stereology...............................................................................................................................29

3.7 WesternBlot...............................................................................................................................30

3.8 Inuteroelectroporation........................................................................................................31

3.9 CellCulture..................................................................................................................................32

3.9.1 Invitroproliferationassays............................................................................................33

3.9.2 Immunocytochemistry......................................................................................................33

3.9.3 Transwellmigrationassay...............................................................................................33

3.9.4 ViabilityAssays.....................................................................................................................33

3.9.5 Electrophysiology................................................................................................................34

4 Results...........................................................................................354.1 EndogenousHCNexpressionduringembryonicdevelopment...........................35

4.2 GenerationofanembryonicHCN‐DNmouseline......................................................37

4.3 MorphologicalcharacterizationofEMX‐HCN‐DNmice...........................................40

4.4 AnalysisofdevelopmentalprocessesintheEMX‐HCN‐DNmouse....................43

4.4.1 ProliferationinEMX‐HCN‐DNmice.............................................................................43

4.4.2 DifferentiationintheEMX‐HCN‐DNmouseline....................................................46

4.4.3 MigrationofcellswithIhablation.................................................................................48

4.4.4 ApoptosisinneonatalEMX‐HCN‐DNmice...............................................................49

4.5 InvitroexperimentsusingprimarycorticalstemcellswithpharmacologicalblockedIh...................................................................................................................................................51

4.6 Identification of HCN‐dependent mechanism in proliferation anddifferentiationofneuralprogenitorcells....................................................................................55

4.7 Summary......................................................................................................................................59

5 Discussion....................................................................................61

Tableofcontent

III

5.1 ExpressionpatternofHCNsubunits................................................................................61

5.2 EMX‐HCN‐DN mouse line as model for HCN channel dysfunction duringembryonicbraindevelopment........................................................................................................64

5.3 TheeffectofIhblockadeonproliferation......................................................................66

5.3.1 RestingmembranepotentialinneuralstemcellsandthecontributionofIh 66

5.3.2 Membrane potential fluctuations during cell cycle progression and thepotentialcontributionoftheIh........................................................................................................68

5.4 Impactofthemembranepotentialonproliferation.................................................70

5.4.1 Calciumsignaling.................................................................................................................70

5.4.2 Directactivationofcellsignalinguponchangesinmembranepotential...71

5.4.3 Cellvolumeregulation.......................................................................................................72

5.5 TheinhibitionofHCNchannelsanditsinfluenceondifferentiation................72

5.6 HCNchannelsinteractionswithotherproteinsandsmallmolecules..............74

5.7 Electrophysiologicalcharacterizationofneuralstemcells...................................76

5.7.1 TheinstantaneousIhcomponent..................................................................................78

5.7.2 SubcellularlocalizationofHCNchannelsubunits.................................................79

5.8 KnownlimitationsofZD7288.............................................................................................81

5.9 Conclusions.................................................................................................................................83

5.10 Outlook.....................................................................................................................................84

6 Supplement.................................................................................866.1 ROIsofStereology....................................................................................................................86

6.2 VectorchartsforIUE..............................................................................................................87

7 References...................................................................................88

8 Listofabbreviations............................................................111

9 Acknowledgement................................................................115

10Erklärung..................................................................................116

Introduction 1

1

1 Introduction

Thecerebraldevelopmentisacomplexprocessrequiringatighttemporalandspatialregulation. Disturbances in any of the underlying processes can lead to severemalformations anddisordersof thebrain, suchasmicrocephaly, lissencephaly, andcortical heterotopia, which can each be associated with epilepsy and mentalretardation(RossandWalsh,2001;Francisetal.,2006;Pangetal.,2008).Tobetterunderstandthepathophysiologyofthesedebilitatingdiseases,againinknowledgeoftheexactmechanismscontrollingcorticaldevelopmentiscrucial.

1.1 CorticalDevelopment

Theformationofthecorticalstructurescanbedividedintothreemajorsteps:

Theproliferationofneuralstemcells;Theirdifferentiationintoneuronsandglialcells;Themigrationofdifferentiatedcellsintocorticallayers.

Cortical pyramidal neurons originate from the ventricular zone (VZ), where thehighestproliferationactivityoccursbetweenembryonicdays(E)10andE17 inthemouse embryo (Takahashi et al., 1993; Molyneaux et al., 2007). They differentiatefromso‐calledneuroepithelialcells,whichsheettheneuraltubeduringearlycerebraldevelopment and give rise to neural progenitor cells, the so‐called radial glial cells(RG)(KriegsteinandGoetz,2003).RGspossessamarkedapicobasalpolarityandarecharacterizedbytheexpressionofglialmarkers(ChoiandLapham,1978;KriegsteinandGoetz,2003).

Two major types of RG progenitor cells (apical and basal progenitors) can bedistinguishedby the location of their somataduringmitosis (Taverna et al., 2014).The apical progenitors proliferate close to the lumen of the ventricle and can bedivided into apical RGs (aRG) and apical intermediate progenitors (aIP) (Tyler andHaydar,2013;De JuanRomeroandBorrell,2015).Thebasalprogenitorsconsistofthe basal RGs (bRG) ‐which exists only in a rather small population in the rodentbrain – and the basal intermediate progenitors (bIP). Both basal cells lose contactwiththeventricleandformthesubventricularzone(SVZ)locatedontheapicalsideoftheVZ(Figure1‐1)(Shitamukaietal.,2011;Wangetal.,2011;DeJuanRomeroandBorrell,2015).

WhileaRGsmaintaintheirabilitytoproliferatethroughoutdevelopmentanddividebothsymmetrically(self‐renewing)andasymmetrically(generatingneuronsor IPs)(Noctor et al., 2004), the IPs and thebRGsalmost exclusivelyperformsymmetrical

1 Introduction

2

neurogenicdivisionsand,thus,divideinaself‐consumingmanner(Haubensaketal.,2004;Noctoretal.,2004;Wangetal.,2011;TylerandHaydar,2013).

In contrast to pyramidal neurons, inhibitory interneurons originate from theganglionic eminence (GE) in the dorsal forebrain. Interneurons exhibit a differenttranscriptionprofile that allows the two typesofneurons tobedifferentiated fromeachotherearly indevelopment (Andersonet al., 1999;Letinic et al., 2002;FishellandRudy,2011).

After thepeakneurogenesis is complete, thegliogenesis starts shortlybeforebirth.Duringbraindevelopment,theSVZgainsinimportancecomparedtotheVZ,whichisreflectedbytheincreasingthicknessoftheSVZandthesimultaneousdecreaseintheVZ,aswellasbythefactthatmostglialprogenitorcells(gIP)proliferateintheSVZ(Figure1‐1).Astrocytesderivefromthreemajorcellpools.OneastrocyticpopulationisgenerateddirectlyfromRGsaftertheirfinaldivisionandtranslocateintotheuppercortical layers to transform into astrocytes (Figure 1‐1) (DeAzevedo et al., 2003;Noctor et al., 2004). Additionally, an overlapping progenitor pool for neuro‐ andgliogenesis is located in the SVZ, which generates astrocytes as early as on E15(Noctoretal.,2004;Gaoetal.,2014;Siddiqietal.,2014).Furthermore,mostof theastrocytes in thepostnatalrodentbrain in layers I ‐ IVaregenerated locallyby thedivision of mitotically active astrocytes (Ge et al., 2012). Oligodendrocytes thatappearearlyonincorticaldevelopmentdescendfromtheGE.FromE12.5toE15.5,theyaregeneratedinthemedialGEandanteriorentopeduncularareaand,atalaterstage,descendfromthelateralandcaudalGE(Kessarisetal.,2006).Asofpostnatalday (P) 0, oligodendrocytes are generated in the SVZ of the dorsal telencephalon(Kessaris et al., 2006). In the adult rodent cortex,most oligodendrocytes originatefrom the lateral and caudalGEor SVZ (LevisonandGoldman,1993;Kessaris et al.,2006).

Introduction 1

3

To reach their predetermined location in the newly formed cortical structure,neuronshavetomigrate.Theearliestbornexcitatoryneuronskeepalongprocesstothebasalmembraneofthepialsurfaceandmigrateoutwardbysomaltranslocationto form the preplate,which is generated at approximately E10 in the rodent brain(Nadarajah and Parnavelas, 2002; Marín et al., 2010). Subsequently born neuronsmigrate via locomotion along these radial glia cells (Rakic, 1972; Nadarajah andParnavelas,2002),therebysplittingthepreplateintothemarginalzoneandsubplate,forming together thecorticalplate (KriegsteinandNoctor,2004).The intermediatezoneisestablishedbytheafferentsofneuronsfromsubcorticalandcorticalregions,building a white matter tract (Rakic, 2009). The cortical plate thickens by newlygenerated neurons in an inside‐out fashion so that neurons born at a later stage

Figure1‐1Processesofcorticaldevelopment

Neuroepithelialcellsintheneuroepithelium(NE)divideandtransformintoradialglialcells.Bothcelltypesextendtheirprocessesthroughthecorticalwalltothemarginalzone(MZ).Apicalradialglialcells (aRG) can divide asymmetrically to generate neurons and symmetrically to establish apicalintermediate progenitors (aIP), the basal radial glial cell (bRG) and basal intermediate progenitor(bIP)pools.bRG,bIPsandaIPsdividemostlysymmetricallyintoneuronsthatmigratealongtheRGsgeneratingthecorticalplate(CP).Aroundbirth,theneurogenesisiscompleteandgliogenesisstarts.Aftertheirlastdivision,RGstransformintoastrocytesandtranslocateintotheCP.OligodendrocytesandfurtherastrocytesaregeneratedbytheglialIPs(gIP)inthethickenedsubventricularzone(SVZ).(VZ:ventricularzone;IZ:intermediatezone);modifiedfrom(KriegsteinandAlvarez‐Buylla,2009).

1 Introduction

4

migrate past their older siblings (Rakic, 1974). The latest generatedneurons reachtheirdestinationin layersII/IIIshortlybeforebirth(Taniguchietal.,2012).Asthecortexthickens,theSVZisformed,andtheVZdecreasesinsizesothatthepostnatalproliferationoccursexclusivelyintheSVZ(Nadarajahetal.,2001).

Cortical interneurons migrate tangentially from the GE along the intermediatezone/SVZorthemarginalzone,followedbyradialmigrationintothecerebralcortex(Andersonetal.,2001;Angetal.,2003).

Once the neurons have reached their destination, contacts with neighboring anddistantneuronsareestablishedbytheformationandpruningofsynapses.

1.2 Electricalactivityindevelopmentalprocesses

The above‐mentioned developmental processes are controlled at numerousmolecular and functional levels. Proliferating, differentiating, and migrating cellsrespondtointernalandexternalcuesguidingthesecellsthroughbraindevelopment.Within an electrically active organ such as the brain, which is composed of cellsequipped with the ability to receive, convert, and transmit electrical signals, thespontaneouselectricalactivityandthebioelectricalpropertiesofthecell'smembraneplayan importantrole inbraindevelopment(Spitzer,2006;RosenbergandSpitzer,2011).Neuronalelectricalactivitydependsontheextrinsicpropertiesofneurons,e.g.neurotransmitter release, and the intrinsic biophysical properties, e.g. ion channelcompositionofthedifferentcelltypes(MoodyandBosma,2005).

Ofnote,electricalactivityisnotonlylimitedtothegenerationofactionpotentialsbutincludestheionfluxthroughthecellmembranemediatedbyavarietyofionchannelsingeneral.

1.2.1 Proliferation

Proliferation isastrictlyregulatedmechanismthatcanbecomepartofpathologicalprocesseswhenitisderegulated.Itiscontrolledbyavarietyofextrinsicandintrinsiccuesmediating cellular responses via, among other factors, secondarymessengers.Calcium is one of the most important second messengers and can modulate cellproliferation in a variety of different cell types. During the cell cycle, calciumtransientsareneededforthetransitionfromG1toSphase,andfortheinitiationandmaintenance ofmitosis (Berridge, 1995; Capiod, 2011). An increase in cytoplasmiccalciumcanbeinducedbyvoltage‐gatedcalciumchannels(VGCC;T‐andL‐type),orbycalciumreleasefromintracellularstores,mainlytheendoplasmaticreticulum(ER)(Capiod,2011).

Introduction 1

5

CalciumtransientsweredetectedinproliferatingcellsintherodentVZinsitu(Owensand Kriegstein, 1998). In those cells, transients are not dependent onneurotransmitters, such as glutamate and γ‐aminobutyric acid (GABA), or onVGCCactivity,butseemtorelyexclusivelyoninternalcalciumstores.Thesefluctuationsincalciumconcentrationssynchronizeinvaryingsizesofcellclustersandinindividualcells(OwensandKriegstein,1998).AnincreaseinintracellularcalciumconcentrationcanalsobedetectedinRGs,inwhichtheinositol1,4,5‐trisphosphate(IP3)pathwayiscoupledtotheinternalcalciumstoresoftheER(Weissmanetal.,2004).Furthermore,inhibitionof thesecalciumtransientsdecreases theproliferationrate(Weissmanetal., 2004; Lin et al., 2007), as reflected by a reduction in S‐phase cells. TheseobservationssuggestaroleforcalciumfluctuationsatthetransitionbetweenG1andS phase (Weissman et al., 2004; Resende et al., 2010).Due to the tight coupling ofproliferatingprogenitorcellsintheVZbygapjunctions,thecalciumwavesleadtoasynchronizationofthecellcycleprogression(Bittmanetal.,1997).

Besides calcium signaling, the proliferation of neuronal precursor cells can bemodified by external cues such as neurotransmitters. GABA and glutamate havediverse roles in the regulation of neural proliferation. Both neurotransmitters canincreaseneuralproliferation in theVZ,while theydecreaseproliferation in theSVZ(Haydar et al., 2000). It is further known that the reduction in proliferation isfacilitated by GABAA and α‐amino‐3‐hydroxy‐5‐methyl‐4‐isoxazolepropionic acid(AMPA)/kainatereceptors,whichmediateadepolarizationofthecellmembranethatleadstoanincreaseinintracellularcalciumthroughVGCCand,consequentially,toadecrease inDNAsynthesis (LoTurcoetal.,1995;Haydaretal.,2000).ThenegativeeffectofGABAontheproliferationofprimarycorticalstemcellswasalsoshownincellcultureexperiments(Antonopoulosetal.,1997).TheprominentinhibitoryeffectofGABAandglutamateonneuralproliferationcancausea substantial reduction inproliferation in the SVZ, which masks the pro‐proliferative effect of theseneurotransmittersintheVZ(Haydaretal.,2000).

Apart from ionotropic receptors, proliferating neuronal precursor cells mainlyexpress calcium‐dependent and voltage‐gated potassium channels (Mienville andBarker, 1997; Hallows and Tempel, 1998; Picken Bahrey and Moody, 2002).Potassiumchannelsarealsoexpressedinotherproliferatingcells,includingdifferentcancer cell lines, and are believed to influence the cell cycle progression bymodulatingthemembranepotential(Blackistonetal.,2009;Urregoetal.,2014).

1 Introduction

6

1.2.2 Differentiation

The differentiation of neural precursor cells into neurons of different fate andmorphologyaswellasintoglialcellsisastrictlyregulatedprocesswithinternalandexternal cues that has a distinct temporal and spatial order (Desai andMcConnell,2000; Rowitch and Kriegstein, 2010). The various differentiation steps are highlysensitivetochangesinionconcentrationinthecellenvironment,anddependontheabilityofthecellstoreacttosuchalterations.

Thedifferentiationofneuronsisdeterminedbytheacquisitionofaneurotransmittertypeand theestablishmentofbiophysicalpropertiesby thedifferentneuron types,mainlybyionchannelcomposition.InembryonicXenopuslaevisneuronsandprimaryembryonicratneurospheres,twodifferentcalciumtransientswereobserved(GuandSpitzer,1995;Ciccolinietal.,2003).On theonehand, theglobal calciumtransientsspreadingthroughthewholecellandmodulating ionchannelandneurotransmittermaturationand,ontheotherhand,calciumevents,whicharemostlylocalprocessesandmodulate neurite outgrowth (Gu and Spitzer, 1995; Ciccolini et al., 2003). ThepatternoftheglobalcalciumtransientscanbecorrelatedtotheneuronalcelltypeinXenopuslaevisspinalcordneurons.Furthermore,suppressionofthisactivityleadstotheexpressionofexcitatoryneurotransmitters(namelyglutamateandacetylcholine),whileanincreaseinthisactivityinducesaninhibitoryneurotransmittertype(glycineand GABA), independent of their genetic fate (Borodinsky et al., 2004). Theseexperimentsshowthattheactivitylevelcanpredominateotherdevelopmentalcues.Also, Brosenitsch and Katz showed that the expression of tyrosine hydroxylase isdependentona combinationof the transcription factorPhox2and thedepolarizingactivity invitro and invivo, suggestinga role for electrical activity indopaminergiccellfate(BrosenitschandKatz,2002).Forallofthesefindings,thereisonlyashorttime window during brain development in which electrical activity plays animportantrole.

In addition to the determination of the neuronal fate, the final morphology withdendritic outgrowth and axonal pathfinding is also regulated by patterns ofspontaneousactivityandcalciumtransients(KatzandShatz,1996;Tangetal.,2003;HansonandLandmesser,2004;Kaliletal.,2011).Furthermore,theestablishmentoftheir axonal and dendritic connections (Spitzer, 2006; Chen and Kriegstein, 2015)and the formation of new synapses (Ben‐Ari, 2006; Wang and Kriegstein, 2009)depend on spontaneous neuronal activity, and is needed for the establishment ofadultpatternsofconnectivitywithindevelopingnetworks(KhazipovandLuhmann,2006).

Although the differentiation of glial cells shows distinct machinery defining theoligodendrocytic or astrocytic cell fate, it also shares somemechanisms among the

Introduction 1

7

different cell types. The calcium‐sensing receptor (CaSR) is upregulated duringoligodendrogenesis in the rodent brain (Ferry et al., 2000; Chattopadhyay et al.,2008). Facilitated by the CaSR, extracellular calcium mediates the commitment ofneural precursor cells to oligodendrocytic precursor cells, their proliferation, andtheirmaturationbyanunknownmechanism(Chattopadhyayetal.,2008).Duringthisprocess, calcium is acting as a primary messenger, instead of assuming itspredominantroleasasecondarymessenger.

Much less is known about the role of calcium in astroglial differentiation. In thepituitary adenylate cyclase‐activating polypeptide (PACAP) pathway, which, uponactivation, is promoting the astrocytic cell fate, calcium is needed as a secondmessenger (Vallejo, 2009). Furthermore, expressionofneuregulin‐1 suppresses theglialcell fatebyapresenilin–dependentpathway. Interestingly,presenilin isknowntohaveafunctionincalciumhomeostasis(HonarnejadandHerms,2012),whichledtothehypothesisthatcalciumwavesinRGsarenotonlyneededfortheproliferationof neurogenic precursor cells but also for the inhibition of the astroglial cell fate(Leclercetal.,2012).

Both glial cells have in common that their proliferating precursor cells expressdelayed rectifying and transient potassium channels, as well as sodium channels(Kressinetal.,1995;BordeyandSontheimer,1997;Chvataletal.,1997;MacFarlaneand Sontheimer, 2000; Higashimori and Sontheimer, 2007). The cell cycle exit isaccompanied by upregulation of the Kir4.1 channel, leading to a hyperpolarizedmembranepotentialand thesubsequentdifferentiation intoglial cells (HigashimoriandSontheimer,2007).

1.2.3 Migration

One modulatory mechanism of the migration of neurons in the rodent brain isdependent on the early secretion of different neurotransmitters. The mechanismsdiffer between the radial and tangential migration of excitatory and inhibitoryneurons,andbetweenthevariousbrainregions.Intherodentcortex,themigrationofneurons is mainly regulated by paracrine secretion of the two majorneurotransmittersglutamateandGABA(Demarqueetal.,2002;Manentetal.,2005).

In migratory pyramidal cells, the main glutamatergic receptor is the N‐methyl‐D‐aspartate (NMDA) receptor (Behar et al., 1999; Reiprich et al., 2005), whereasinterneuronsexpressmainlyAMPA/kainate receptors (ManentandRepresa,2007).Forradiallymigratingexcitatoryneurons,glutamateisachemoattractantmediatingcellmigrationtowardstheouterlayerofthecorticalplate(Beharetal.,1999;Hiraietal., 1999; Reiprich et al., 2005). Compared to theirmature siblings, themembranepotentialof immaturedissociatedcerebellarneuronsisinamoredepolarizedstate.

1 Introduction

8

This depolarization removes themagnesium block of NMDA receptors, while theirsubsequent activation mediates a calcium influx (Hirai et al., 1999; Gerber et al.,2010).

TheroleofGABAinthemodulationofmigrationofcorticalneuronsismorecomplexthan thatof glutamate.While theearlypost‐mitoticneurons in theVZ/SVZexpressthe ionotropic GABAA and GABAA/C receptors ‐ which promotemigration from thegerminal zones into the IZ ‐ neurons migrating through the IZ react via themetabotropicGABABreceptor,which isneeded toenter thecorticalplate (Beharetal.,1998,2000).Intheuppercorticalplate,neuronsexpressGABAAreceptors,whichmediateastopsignaland,concomitantly,adecreaseinthemigrationrate(Beharetal., 1998, 2000). GABA‐evokedmigratory responses are also dependent on calciuminflux (Heck et al., 2007). In mature neurons, GABA is the main inhibitingneurotransmitterintheforebrain,whileinimmatureneuronsGABAisdepolarizing.The depolarizing GABA effect is due to an increased intracellular chlorideconcentration in the immature brain, mediated by a differential expression ofchloride transporters (Ben‐Ari, 2002). It has been shown that this so‐called GABAswitch is needed in interneurons to alter theGABA response fromapro‐migratorysignaltoastopsignal(BortoneandPolleux,2009;Inoueetal.,2012).

Furthermore,serotoninwasidentifiedasaninhibitorofradialmigration(Riccioetal.,2012).

Both glutamate and GABA increase the intracellular calcium concentration, whichstronglyinfluencesmigration(Beharetal.,1996;Hiraietal.,1999).Itisknownthatthecalciuminfluxingranulecellsofthecerebellumismediatedbythevoltage‐gatedN‐type calcium channels. The amplitude and frequency of these calcium transientsare correlated to the mode of migration, with a high intracellular calciumconcentrationpromotingthemigration(KomuroandRakic,1996).

In addition to calcium channels, small sodium currents and delayed potassiumcurrents, but no hyperpolarization‐activated currents, were detected in migratingneurons of the murine cortex (Picken Bahrey and Moody, 2002). By ablating thefunctionof the leakpotassiumcurrentmediatedby theKCNKchannel,migrationofthe pyramidal cells is impaired (Bando et al., 2014). This ablation leads to alteredcalciumhandlinginthemigratingcells(Bandoetal.,2014).

Introduction 1

9

1.3 HCNchannels

Asdescribedabove,earlyelectricalactivityisofgreatimportancetoallprocessesofcortical development. For a developing neuron to be responsive and to processelectrical signals, the intrinsic biophysical properties are essential. Thehyperpolarization‐activated cyclic nucleotide‐binding non‐selective cation (HCN)channelisofhighrelevancetothedeterminationoftheseproperties.

1.3.1 StructureandfunctionofHCNchannels

The HCN channel is a tetrameric channel belonging to the superfamily of voltage‐gated potassium channels (Ludwig et al., 1998; Kaupp and Seifert, 2001). Eachsubunitconsistsofsixtransmembranedomains(S1‐S6)withS4asavoltagesensorcontainingpositivelychargedarginineandlysineresidues(Figure1‐2)(Gaussetal.,1998; Ludwig et al., 1998).Thepore region is locatedbetweenS5 andS6with theglycine–tyrosine–glycine(GYG)motif,whichischaracteristicoftheselectivityfilterinpotassiumchannels.Incontrasttostrictlypotassium‐conductingchannels,theHCNchannelcontainsacysteineresidueinfrontoftheGYGmotifspacedbyanisoleucine.This is presumably the cause of the additional sodium permeability of the channel(Macrietal.,2012).AttheintracellularC‐terminus,aC‐linkerconnectstheS6domainwiththecyclicnucleotide‐bindingdomain(CNBD)(Figure1‐2)(Waingeretal.,2001;Zagotta et al., 2003). HCN channels can assemble into homo‐ or heteromerictetramers(Figure1‐2)(UlensandTytgat,2001;Muchetal.,2003).

Figure1‐2HCNchannels

EachHCNsubunitconsistsofsixtransmembranedomains(S1–S6)withapositivelychargedvoltagesensorinS4.TheporeregionislocatedbetweenS5andS6withtheGYGmotifasaselectivityfiltercharacteristic of potassium channels. At the C‐terminus, the cyclic nucleotide‐binding domain(CNBD) is connected via the C‐linker to S6, mediating the cyclic nucleotide modulation. Foursubunitsformahomo‐orheteromerictetramer.

1 Introduction

10

In both the rodent and the human genomes, four different genes encode HCNsubunits(HCN1to4)(Ludwigetal.,1998;Santoroetal.,1998;Stieberetal.,2005).The distinct subtypes show specific kinetics,with theHCN1 channel exhibiting thefastestactivation(Santoroetal.,1998),HCN2andHCN3beingintermediateintheirresponse (Ludwig et al., 1999; Stieber et al., 2005), and HCN4 being the slowestreactingsubtype(Ludwigetal.,1999;Seifertetal.,1999).Inadditiontothedifferentactivationanddeactivationdynamicsofthechannelsubtypes,theresponsetocyclicnucleotides varies aswell. HCN2 and HCN4 subtypes show the strongest effect oncAMPapplication (Ludwigetal., 1999;Seifert et al., 1999),whileHCN1reactsonlyslightlytocAMP(Santoroetal.,1998),andHCN3showsnoeffectofcyclicnucleotidebinding(Stieberetal.,2005).

The function ofHCN channels and the h‐current (Ih)mediatedby them, is stronglydependentonthespatialdistributionoftheHCNsubunitsinneuronalcompartments,whichsubstantiallyvariesbetweenindividualneuronalsubtypes.

In the cortex and hippocampus, HCN channels are mainly expressed in distaldendrites (Lörincz et al., 2002), where they contribute to the integration ofpostsynapticpotentialsand,thus,totheregulationoffunctionalconnectivity(Magee,1999;WilliamsandStuart,2000;Huangetal.,2009).

Inavarietyofcelltypesindifferentbrainregionssuchasthehippocampus,thalamus,andcerebellum,HCNchannelsaremainlydistributedalongthesomaandpresynapticterminalsofaxons(Lujánetal.,2005;Boyesetal.,2007),influencingthemembraneproperties and neurotransmitter release (Lupica et al., 2001; Ludwig et al., 2003;Boyesetal.,2007).Also,Ihcancontributetothefrequencycontrolofthepacemaker,duetoitsdepolarizingactionuponhyperpolarization(BrownandDifrancesco,1980).This was shown in the thalamus and in the heart (Brown and Difrancesco, 1980;McCormickandPape,1990;Mesircaetal.,2014).

In heterologous systems and in some cell types such as stellate cells of ventralcochlearnucleus,theIhcanbeseparatedintoaninstantaneouscurrentIins,whichisvoltage‐independent, and a steady‐state current Iss, which is slowly activated uponhyperpolarization (Macri et al., 2002; Proenza et al., 2002; Rodrigues and Oertel,2005;Aponteetal.,2006).ItisnotyetknownwhethertheIinsisaleakconductanceorasecondchannelpopulation(Macrietal.,2002;ProenzaandYellen,2006).

1.3.2 ExpressionpatternsofthedifferentHCNsubtypes

WhileIhhascrucialfunctionsintheheartandbrain,thisworkwillmainlybefocusedonthelatter.

Introduction 1

11

AllfourHCNsubtypescanbefoundwithdifferentexpressionpatternsintherodentbrain,whichwasinvestigatedattheRNAandproteinlevels(Moosmangetal.,1999;NotomiandShigemoto,2004).ThehighestexpressionofHCN1canbe found in theneocortex,hippocampus,superiorcolliculus,cerebellum,andbrainstem(Moosmangetal.,1999;Santoroetal.,2000;Lörinczetal.,2002;NotomiandShigemoto,2004).TheHCN2 subtype is ubiquitously expressed throughout the brainwith prominentlabelinginthethalamusandbrainstem(Ludwigetal.,1998;Moosmangetal.,1999;Santoro et al., 2000;Notomi andShigemoto, 2004).Of all theHCN subtypes,HCN3exhibitstheweakestexpressioninthebrain,withonlyaslightincreaseinexpressionin the olfactory bulb, piriform cortex, habenular nucleus, and hypothalamus(Moosmang et al., 1999; Notomi and Shigemoto, 2004). HCN4 has a distinctexpressionpatternwithprominentlabelingintheolfactorybulb,habenularcomplex,andsubstantianigra,andwiththehighestexpressioninthethalamus(Moosmangetal.,1999;Seifertetal.,1999;Santoroetal.,2000;NotomiandShigemoto,2004).

1.3.3 RegulationofHCNchannels

HCN‐channelactivityisregulatedbydifferentmolecules,kinases,andotherproteins.

Thebest‐investigatedregulatorsarethecyclicnucleotides,mainlycAMPand,withalower binding affinity, cGMP. Upon binding of cyclic nucleotides to the CNBD, theactivationcurveshifts towardsmoredepolarizedmembranepotentials leading toamore positive half‐maximal activation. This effect is caused by the release of auto‐inhibition, probably mediated by the C‐linker (Figure 1‐2) (Wainger et al., 2001;Wicks et al., 2010). Additionally, cAMP traps the channel in the open state, whichresults not only in faster activation but also in slower deactivation (Wicks et al.,2010).ThesmalleffectofcAMPbindingtoHCN1isduetoapre‐existingdisinhibitionof the subunit (Chowet al., 2012).Another smallmoleculewith the sameeffect onHCNchannels, i.e. adepolarizationshiftof theactivationcurveandanalteration inchannelkinetics,ismediatedbyphosphatidylinositol4,5‐bisphosphate(PIP2)(Pianetal., 2006;Zolleset al., 2006).PIP2 acts independentlyof cAMPandshows the sameeffectonHCN1,2,3,and4(Pianetal.,2006;Zollesetal.,2006;Yingetal.,2011).Bothmolecules lead to an activation of Ih at physiological membrane potentials.Furthermore,intracellularprotonsshifttheactivationcurvetomorehyperpolarizedpotentialsandcauseadecreaseinchannelopening(MunschandPape,1999;Zongetal.,2001).ThisresultsinareductionofIhduetoacidificationandtoanincreaseinthecurrentfacilitatedbyalkalizationoftheintracellularmilieu(MunschandPape,1999;Zong et al., 2001).Moreover, Ih ismodulated by extracellular protons in amannersimilartothatobservedwithintracellularacidification.Expressedintastebuds,HCNchannels function as a receptor for sour taste (Stevens et al., 2001). In addition, adecrease in the extracellular chloride concentration reduces Ih and decelerates theopeningkinetics inHCN2andHCN4channels(McCormickandPape,1990;Fraceet

1 Introduction

12

al.,1992;Wahl‐Schottetal.,2005).Interestingly,areductioninintracellularchlorideconcentration increases the instantaneous component without affecting the steadystateIss(Mistríketal.,2006).

In addition to small molecules, phosphorylation of kinases can also regulate HCNchannels.HCN1wasdiscoveredbythebindingtoaSrctyrosinekinase(Santoroetal.,1997). Later itwas shown that Src kinasephosphorylationat theC‐linker inHCN2andHCN4shiftstheactivationcurvetomorepositivepotentialsandacceleratestheactivation kinetics (Zong et al., 2005; Arinsburg et al., 2006; Li et al., 2009). Thephosphorylationsiteisconservedinallfoursubtypes.

Like other ion channels, the function of HCN channels is dependent on the correctsubcellularlocalization.Therearedifferentaccessorysubunitsandproteinsthatareneeded to determine the final localization of the HCN subunits. Some exhibitadditional regulatory functions, but all interact via the C‐terminus of the channelsubunit.HCN1interactswithfilaminA.Uponbinding,thechannelformsclustersandcanbeendocytosed,resultinginadecreaseincurrentdensity(Gravanteetal.,2004;Noam et al., 2014). Additionally, there is a deceleration of channel kinetics and ahyperpolarization shift of the activation curve (Gravante et al., 2004). Frominteraction studies, it is known that HCN2 can bind to the neuronal scaffoldingproteinstamalin,S‐SCAM,andMint2(Kimuraetal.,2004).Thesescaffoldingproteinsare essential for trafficking and localization of several membrane proteins.Furthermore,HCN2interactswiththepotassiumchannelregulatorprotein1(KCR1),whichresultsinasuppressionofchannelactivityintheheart(Michelsetal.,2008).Inthe rodent brain, HCN3 interacts with the potassium channel tetramerizationdomain‐containing protein 3 (KCTD3), which increases the HCN3 cell surfaceexpressionand,thereby, itscurrentdensity(Cao‐Ehlkeretal.,2013).Aninteractionwithcaveolin‐3isknownforHCN4inthesinoatrialnodeoftheheart(Barbutietal.,2004). This interaction places the channel into lipid rafts, facilitating proximity tomodulating proteins such as adrenergic receptors (Barbuti et al., 2004, 2007). It isopen to speculationwhether thedetecteddepolarizing shift of the activation curvecausedbythedestabilizationofthelipidraftsisduetothedisruptedinteractionwithcaveolin‐3, or due to hitherto unknown proteins in the rafts (Barbuti et al., 2004).HCN1,2,and4interactwiththeMinK‐relatedprotein1(MiRP1)orKCNE2,resultinginanaccelerationandincreaseofIh,aswellasinahighercellsurfaceexpressionofthe protein in the heart and brain (Brandt et al., 2009; Ying et al., 2012). ThetetratricopeptiderepeatcontainingtheRab8b‐interactingprotein(TRIP8b)interactswithall fourHCNsubtypesandregulates traffickingandgatingof theHCNchannel(Santoro et al., 2004, 2011). To date, nine splice variants of TRIP8b with variousbindingeffectsonthechannelareknowntoresultintheup‐ordownregulationofthecellsurfaceexpressionindistinctcellularlocations(Lewisetal.,2009;Santoroetal.,

Introduction 1

13

2009). In addition, TRIP8b antagonizes cAMP binding to the C‐terminus of thechannelsubunit,resultinginanegativeshiftoftheactivationcurveandinareducedchannelopeningprobability(Santoroetal.,2009;Zollesetal.,2009;Huetal.,2013;DeBergetal., 2015). It is stillunknownwhether this is theonly impacton channelgating or whether TRIP8b also has a cAMP‐independent mechanism (Zolles et al.,2009).

1.3.4 KnockoutmousemodelsofthedifferentHCNsubunits

Knockout models of all four HCN subtypes are currently being analyzed. TheirphenotypesvaryduetothedistinctexpressionprofilesofthevariousHCNsubtypes.

A global HCN1 knockout shows an increase in seizure susceptibility and severity,independentofthe inductionprotocol,withoutdisplayingspontaneousepileptiformactivityinthemurinebrain(Huangetal.,2009;Santoroetal.,2010).Also,thesemiceshowenhancedhippocampus‐dependent learningandmemorybut impairedmotorlearning(Nolanetal.,2003).Furthermore,ananti‐depressanteffectwasobservedinbehavioral testswithHcn1knockoutmice(Lewisetal.,2011). In thevisualsystem,HCN1iswidelyexpressedintheretina.Upongeneticablation,theretinalresponsetolightisprolonged,leadingtoalteredvision(Knopetal.,2008).Moreover,becauseoftheexpressionofHCN1channelsindorsalrootganglia,theirroleinneuropathicpainwas investigated.Hcn1knockoutmiceshowadecrease incoldallodynia(Mominetal., 2008). In addition, HCN1 knockout mice exhibit bradycardia with sinusarrhythmiaandsinuspauses(Fenskeetal.,2013).

InlinewiththeexpressionpatternofHCN2,Hcn2knockoutmiceshowspontaneousabsence epilepsy with reduced locomotion and sinus dysrhythmia (Ludwig et al.,2003). Likewise, a spontaneously occurringmutation in theHcn2 gene resulting inthe absence of the HCN2 subtype causes absence epilepsy, ataxia, and an anti‐depressanteffectinso‐calledapatheticmice(Chungetal.,2009;Lewisetal.,2011).DuetotheseverityoftheglobalknockoutofHcn2,thegeneticdeletioninasubtypeofsomatosensory neurons was established and investigated, showing a decrease ininflammatorythermalhyperalgesiaandneuropathicpain(Emeryetal.,2011).

The knockout of the auxiliary subunit TRIP8b shows deficits inmotor learning, ananti‐depressanteffect,andmildabsenceepilepsy(Lewisetal.,2011;Heuermannetal.,2016).

GlobalHcn3knockoutmicepresumablyhaveacardiacphenotypewithanincreaseinT waves in the electrocardiogram, indicating a decelerated cardiac repolarization(Fenskeetal.,2011).

1 Introduction

14

TheglobalknockoutofHcn4 leadstoanembryonically lethalphenotype,duetotheabsenceofthemaincardiacpacemaker,andcausesseverebradycardia(Stieberetal.,2003).

1.3.5 HCNchannelsanddisease

AlthoughalterationsinHCNchannelsandIhareknowntobeinvolvedinavarietyofphysiological and pathological processes in the nervous system, such as in pain(LewisandChetkovich,2011)andParkinson’sdisease(Chanetal.,2011),thisworkwillbefocusedontheimpactofIhonepilepsy.

In different animal models of genetic and induced epilepsies, e.g. febrile seizures,temporal lobeepilepsy,orabsenceepilepsy,Ihwasshowntobeeither increasedordecreased,dependingonthetemporalorregionalspecificityoftheexperiments(see1.3.4)(Chenetal.,2001;Jungetal.,2007;Dyhrfjeld‐Johnsenetal.,2009;Kanyshkovaetal.,2012).

Inhumanswithseverechronicepilepsy,changescanbeobservedinHCNexpressionand current density in the hippocampus and neocortex (Bender et al., 2003;Wierschke et al., 2010). In addition to the alteration in Ih, which is likely to besecondary to epilepsy, genetic variations and triple proline deletions in HCN1 andHCN2weredetectedinpatientswithgeneralizedseizuresorfebrileseizures(Tangetal.,2008;Dibbensetal.,2010).Recently,mutationsintheHCN1subtypeandavarietyof comorbidities were found in patients with epileptic encephalopathies, whichaltered the HCN1 properties expressed in vitro in heterologous cells (Nava et al.,2014).ThesevereneurologicalphenotypesassociatedwithHCN1mutationspointtoanimportanteffectofIhonbraindevelopment.

Intheheart,HCNchannelsandIhcanbechangeddependingontheprogressionofthedisease,ortheycanconstituteageneticriskfactor for theonsetof thedisease(see1.3.4)(Herrmannetal.,2015;VerkerkandWilders,2015)

1.3.6 HCNexpressionandfunctioninthedevelopingrodent

Depending on the neuronal activity, changes occur in HCN subunit expression andlocalizationinthecellduringpostnataldevelopment.Theadjustmentoftheneuron'sbiophysical properties is an essential prerequisite for the establishment andmaturationofnetworks(VasilyevandBarish,2002;Brewsteretal.,2006;Surgesetal., 2006;Bender andBaram, 2008).However, the prenatal impact of Ih on corticaldevelopmentremainsunknownandhasyettobeelucidated.

In the developing heart, HCN channel‐mediated pacemaker currents are crucial forcardiacdevelopmentand,thereby,fortheviabilityoftheembryo(see1.3.4)(Stieber

Introduction 1

15

etal.,2003).Thesamelethalitywasobservedinmicethatweredeficientinthecyclicnucleotide‐binding site in the CNBD of the HCN4 subtype, suggesting that theenhancement of HCN4 activity due to cAMP binding is an essential feature in thedevelopment of the cardiac system (Harzheim et al., 2008). Surprisingly, Hcn4deletion with an expression onset in the adult mouse shows only a mild cardiacphenotype with no alteration in mortality (Herrmann et al., 2007), suggesting animportantrolefortheHCN4subtypeincardiacdevelopment.

Ofnote,HCNexpressionwasalreadydetectedinmurineembryonicstemcells(Wanget al., 2005). While the impact of HCN channels on stem cell proliferation wasdescribed, themechanisms underlying disturbed cell cycle regulation remain to beexplored(Lauetal.,2011).

1.4 Hypothesis

ThisprojectisaimedatinvestigatingthefunctionofIhintheprenataldevelopmentofthemurine cortex based on a transgenic approach with a dominant‐negative HCNsubunit, ablating the Ih in a subunit‐independent manner. To evaluate the earlyembryonicdevelopment,theEMX1promoterwasusedtoexpressthetransgeneinaforebrain‐specific manner from E10 onward. This approach possibly altered thebiophysicalpropertiesoftheneuronalprecursors,sheddingnewlightontheimpactof electrical competence on early developmental processes. Additionally, in uteroelectroporation was applied to further dissect specific prenatal time periods andprocesses. The combined evidence of these in vivo data and in vitro experiments ‐usingratcorticalstemcellculturesandblockingIhpharmacologicallywithZD7288–showtheroleofIh intheearlydevelopmentofneuronalstemcells,progenitorcells,andearlyneurons.

This data provide novel insight into the impact of HCN channel activity on earlyprenataldevelopmentalprocessesofthebrain.

2 Material

16

2 Material

2.1 Substances

AllchemicalsusedinthisworkarefromCarlRothorSigma‐Aldrichifnototherwisestated.

5‐Bromo‐2’‐deoxyuridine(BrdU);invitro Fluka

Betaisodona mundipharm

Bovineserumalbumin PAA

Buprenorphine RBPharmaceuticals

Caprofen NorbrookLaboratories/Bayer

Chemiblocker MerckMillipore

DAPIFluormount‐G SouthernBiotech

DMEM/F‐12medium Gibco

dNTPS PANBiotech

Fluormount SouthernBiotech

Gadolinium Insightagents

Hoechst33342 Invitrogen

Humanrecombinantfibroblastgrowthfactor(FGF) Invitrogen

Isofluran AbbVie

Ketamin Albrecht

L‐glutamine,N2‐supplement Gibco

Luminata™CrescendoWesternHRPSubstrate MerckMillipore

Lysisbuffer(genotyping) Peqlab

MidoriGreenAdvance Biozym

Milkpowder Heirler

NeuroTrace®500⁄525 MolecularProbes

NeuroTrace®530⁄615 MolecularProbes

Normalgoatserum Vector

NuPAGE®MOPSSDSrunningbuffer Lifetechnologies

Material 2

17

NuPAGE®transferbuffer Lifetechnologies

OCT Sakura

PANladder1(DNAmarker) PANBiotech

Penicillin/Streptomycin Gibco

PicrinicAcid AppliChem

Sodiumpyruvate Gibco

Spectra™Multicolorbroadrangeproteinladder Fermentas

TaqPolymerase AppliedBiosystems

TrispH7.4 AppliChem

TritonX100 USBiological

Tween20 AppliChem

Vidisic BauschundLomb

Xylazin Eurovet

2.2 Buffers

Bathsolution

135mMNaCl5mMKCl 5mMHEPES 2mMCaCl2x2H2O 2mMMgCl2x6H2O 10mMGlucose 20mMSucrosepH7.4(NaOH)

Carriersolution1%HorseSerum0.2%BSA0.5%TritonX‐100

Electrodesolution

5mMKCl 135mMK+‐Gluconate 10mMHEPES 0.2mMNaGTP 2mMMgCl2x6H2O 2mMATP(Mg‐Salt)0.1mMEGTApH7.3(KOH)

HCN‐IHCsolution5%Chemiblocker0.5%TritonX‐100inPB

2 Material

18

LysisBuffer(genotyping) 0.4mg/mlproteinaseKinlysisbuffer

PCRbuffer(10x) 0.2MTrispH7.90.5MKCl

Phosphatebuffer(PB)0.1M14.41gNa2HPO4x2H2O2.62gNaH2PO4xH2Oad1lpH7.4

Phosphatebufferedsaline(PBS)3MNaCl161mMNa2HPO4x2H2O39mMKH2PO4

Sodiumcitratebuffer 10mMSodiumcitrateinPBSpH9(NaOH)

Sucrosebuffer

0.32MSucrose1mMEDTA5mMTrisHClpH7.4+ freshly added Proteinase InhibitorCocktail(1:100)

Trisbufferedsaline(TBS) 100mMTrispH7.450mMNaCl

2.3 Antibodies

2.3.1 PrimaryAntibodies

Antigen Clone Species Producer

WB

HCN1 N70/28 mouse neuromab

HCN2 75‐111 mouse neuromab



Hemagglutinin(HA) 12CA5 mouse Roche

IHCandICC

CleavedCaspase3 AF835 rabbit RandDSystems

CNPase 11‐5B mouse Sigma

GFAP GA5 mouse Millipore

HCN4 PG2‐1AY rat AGKaupp/Wachten

Material 2

19

Ki67 AB15580 rabbit Abcam

Nestin rat‐401 mouse Millipore

NeuN A60 mouse Chemicon

TuJ1 TuJ‐1 mouse RandDSystems

BrdU(ICC) BU‐33 mouse Sigma‐Aldrich

Sox2 MAB2018 mouse RandDSystems

2.3.2 Secondaryantibodies

Antigen Conjugate Species Producer

GFP A488 Rabbit MolecularProbes

Goat A546 Donkey MolecularProbes

Mouse HRP Sheep Dianova/Jackson

Mouse A488 Goat MolecularProbes

Mouse A546 Goat MolecularProbes

Mouse fluorescein Goat Invitrogen

Rabbit A546 Goat MolecularProbes

Rabbit fluorescein Goat Invitrogen

Rat A488 Goat MolecularProbes

Rat A546 Goat MolecularProbes

2.4 PrimerSequences

2.4.1 Genotyping

Cre 5’‐AAACGTTGATGCCGGTGAACGTGC‐3’

5’‐TAACATTCTCCCACCGTCAGTACG‐3’

HCN4 5‘‐GGCATGTCCGACGTCTGGCTCAC‐3‘

5’‐TCACGAAGTTGGGGTCCGCATTGG‐3’

2 Material

20

tTA2 5’‐CCATGTCTAGACTGGACAAGA‐3’

5’‐CTCCAGGCCACATATGATTAG‐3’

Ai9 5’‐AAGGGAGCTGCAGTGGAGTA‐3’

5’‐CCGAAAATCTGTGGGAAGTC‐3’

5’‐GGCATTAAAGCAGCGTATCC‐3’

5’‐CTGTTCCTGTACGGCATGG‐3’

NEXCre 5’‐CCGCATAACCAGTGAAACAG‐3’

5’‐GAGTCCTGGAATCAGTCTTTTTC‐3’

5’‐GAGTCCTGGAATCAGTCTTTTTC‐3’

2.5 Kits

BCAassay ThermoFisher

CytoSelectTM24‐WellCellMigrationAssay CellBiolabs

EndoFreePlasmidMaxiKit Quiagen

GeneUpTotalRNAMiniKit Biotechrabbit

KAPASYBRFASTuniversalKit Peqlab

LDHCytotoxicityAssayKit ThermoScientific

LIVE/DEADCell‐MediatedCytotoxicityKit Lifetechnology

QuantiTectReverseTransriptionKit Quiagen

TissueScanDevelopmentalMouseTissueqPCRArray Origin

Methods 3

21

3 Methods

3.1 Animalsandhusbandry

All experimental procedures were approved by the „Behörde für Gesundheit undVerbraucherschutz“of theCityStateofHamburg,andby the„Landesamt fürNatur,UmweltundVerbraucherschutzNordrhein‐Westfalen“,Germany.Animals(mice,Musmusculus)werekeptintypeII longplasticcagesunderstandardhousingconditions(21±2°C,50‐70%relativehumidity,andfoodandwateradlibitum,nestingmaterialas well as a cart box houses andwood for chewingwere provided in the colognefacilities).Micewerekeptonaninverted12:12light:darkcycleandfedwithSniff,LasVivendis or Altromin food. In experiments with doxycyclin treatment 200 mg/kgdoxycyclinewasincludedintotheSnifffood.Transgenicmouselinesused,werethefollowing

C57BL/6‐Emx1tm1(cre)Ito ROSA26tm(tTA) Tg(tetO‐Hcn4GYS,‐EGFP) further on called EMX‐HCN‐DN,

C57BL/6‐Emx1tm1(cre)ItoGt(ROSA)26Sortm9(CAG‐tdTomato)HzefurtheroncalledEMX‐Ai9,

C57BL/6‐ Emx1tm1(cre)Ito ROSA26tm(tTA) Tg(tetO‐Hcn4573X,‐EGFP) further called EMX‐HCN‐573X,

C57BL/6‐Neurod6tm1(cre)KanROSA26tm(tTA) Tg(tetO‐Hcn4GYS,‐EGFP) further calledNEXCre‐HCN‐DN.

NEXCre mice were a kind gift from Prof. Klaus‐Armin Nave, Max‐Planck‐InstituteGöttingen.Aswild‐typemiceC56Bl/6Jwereanalyzed.

3.2 Genotypingoftransgenicmouselines

To genotype the different transgenicmice, polymerase chain reactions (PCR)wereperformed.Tailorearbiopsieswerelysedovernightin100µllysisbufferat54°C.Toinactivate the added proteinase K, samples were incubated at 85 °C for 45 min.Thereafter,aPCRwasperformedusingaBiometraT3Thermocycler.

ThedifferentPCRstogenotypethedifferentmouselinescanbeseeninTable3‐1.

3 Methods

22

Table3‐1PerformedPCRsforgenotypingtheusedmouselines

Forallprimersequencesseethematerialsection.TovalidatetheexistenceoftheCrelocus in the different EMX1 driven mouse lines the following PCR assay withoutdimethylsulfoxide(DMSO)wasperformed:

Forwardprimer(50pmol) 0.5µlReverseprimer(50pmol) 0.5µlMgCl2(50mM) 2.0µl10xPCR‐Buffer 5.0µldNTPs 0.5µlTaq‐Polymerase 0.5µlDNA‐Template 2.0µlH2O ad50µl

InTable3‐2theoptimalPCRprogramtodetecttheCrelocusisshown.

Temperature[°C] Time Cycle

Initialdenaturation 95 5min 1

Denaturation 95 30s

2‐30Annealing 55 45s

Elongation 72 45s

Finalelongation 72 5min 31

Table3‐2PCRprogramusedtodetecttheCretransgene

Mouseline PCR

EMX1‐HCN‐DN Cre,tTA2,HCN4

NEXCre‐HCN‐DN NEXCre,tTA2,HCN4

EMX‐Ai9 Cre,Ai9

EMX‐HCN‐SND Cre,tTA2,HCN4

Methods 3

23

ThetTA2,HCN4,NEXCre,andAi9lociweregenotypedwiththefollowingPCRassaysupplementedwithDMSO:

Forwardprimer(50pmol) 0.5µlReverseprimer(50pmol) 0.5µlMgCl2(50mM) 2.0µlDMSO 2.5µl10xPCR‐Buffer 5.0µldNTPs 0.5µlTaq‐Polymerase 0.5µlDNA‐Template 2.0µlH2O ad50µl

To determine, if the genotype is hetero – or homozygous theNEXCre andAi9 PCRassayswereperformed,usingfourprimersforAI9andthreeprimersforNEXCre.Seematerialsectionfordetail.

ThePCRprogramyieldingthehighestamplificationforthetTA2transgeneisshowninTable3‐3.



Denaturation 95 30s


Elongation 72 45s


Table3‐3PCRprogramtodetectthetTA2transgene

FortheHCN4transgenetheoptimalPCRprogramisshowninTable3‐4.



Denaturation 95 30s


Elongation 72 30s


Table3‐4PCRprogramtovalidatetheHCN4transgene

TodetecttheNEXCrelocusthePCRprogramshowninTable3‐5wasused.

3 Methods

24

Table3‐5PCRprogramtodetecttheNEXCrelocus

The tdTomato in the Ai9 mouse line was detected by the PCR program shown inTable3‐6.



Denaturation 95 30s


Elongation 72 45s


Table3‐6PCRprogramtovalidatetheAi9genotype

TodetectthePCRproductanagarosegelelectrophoreseswasperformedusing2–3% agarose gels with 0.004 % Midori Green Advance (Biozym) to detect doublestrandedDNA.ThesizesofthedetectedDNAbandsforthedifferentgenotypeswerecomparedwiththemarkerPANladder1andareshowninTable3‐7.

PCRproduct Size[bp]

Cre 214

tTA2 597

HCN4 350

NEXCre WT770,Mut520

Ai9 WT297,Mut196Table3‐7PCRproductsandtheirsizesinbasepairs[bp]



Denaturation 95 30s


Elongation 72 30s


Methods 3

25

3.3 MRIAnalysis

ForthemagneticresonanceimagingneonatalanimalsweredeeplyanesthetizedwithaKetamin(100mg/ml)andXylazin(20mg/ml)solutionandperfusedwith7mlPBSand4%paraformaldehyde(PFA)inPBS,bothsolutionscontained3%gadoliniumasMRIcontrastagent.The totalheadwas fixatedovernight in theabovementionPFAgadoliniumsolutionandstoredin3%gadoliniuminPBSuntilthemeasurement.TheMR imagingwas performed in a 3.0 T small animal scanner in a 3Dmeasurement(ClinScan,Bruker).Neonatalanimalswereanalyzedwithturbospinecho(repetitiontime(TR)200ms,echotime(TE)8.5s,turbofactor8,matrix144x192pixel,fieldofview(FOV)11x14mm,slicethickness80µm).3weekoldanimalsweremeasuredaliveunder1‐2%Isoflurananesthesia(500ml/mininair).Thebreathingrateandbody temperature of the head fixed animals was monitored. The juvenile animalswereMRimagedbyaconstructive interference insteadystate(CISS)sequence(TR8.14ms,TE4.07s,matrix192x192pixel,FOV16x16mm,slicethickness90µm).AllfurtheranalyseswereperformedwithOsirixv.3.9.4.software32bitversion.

3.4 QuantitativerealtimePCR

To analyze the expression levels of the different HCN subtypes during themurinedevelopment of the brain theTissueScanDevelopmentalMouseTissueqPCRArrayfromOriginwasperformed.Thedetaileddescriptionofthemethodcanbefoundin(Schlusche, 2011). In short, theTaqManprobesof theHCN1 toHCN4 subtypes arelabeled with the VIC™ fluorophore while the GAPDH is marked with FAM™fluorochrome,toenabletheexecutionofaduplexPCR.Allprimersandprobeswereobtained from Applied Biosystems (HCN1: Mm00468832m1; HCN2:Mm00468538m1;HCN3:Mm01212852m1;HCN4:Mm01176086m1).ToamplifythePCR‐productsastandardprotocolshowninTable3‐1Table3‐8byutilizinga7900HTFastReal‐TimePCRSystemfromAppliedBiosystemswasperformed.


Enzymeactivation 50 2min 1

Initialdenaturation 95 10min

2‐40Denaturation 95 15s

Synthesis 60 1min

Table3‐8qPCRprotocolusedtoinvestigatetheexpressionlevelofthedifferentHCNsubtypesduringdevelopment

3 Methods

26

Analysiswasperformedusing theSDS2.3program tocalculate the cycle threshold(CT) values. All further analyseswere processed by the RQManager 1.2 and DataAssistv2.0software.

3.5 Nextgenerationsequencingincludingsamplepreparation

Thenextgenerationsequencing(NGS)resultswereobtainedandkindlyprovidedbyDr. Marta Florio of the laboratory of Prof. Huttner at the Max‐Planck‐Institute inDresden,Germany.ForthedetailedprotocolseeFlorioetal.,2015.

In short, cells were sorted using fluorescence‐activated cell sorting (FACS) infollowinggroups:

forthemouseembryo:apicalradialgliacells(aRG):TUBB3‐GFP‐/DiI+/Prom1+;basalradialgliacells(bRG):TUBB3‐GFP–/DiI+/Prom1–;basalintermediatedprogenitorcells(bIP):TUBB3‐GFP–/DiI–/Prom1–;neurons(N):TUBB3‐GFP+/DiI+/Prom1–.Forthehumandataset:aRG:DiI+/Prom1+/S‐G2‐MbRG:DiI+/Prom1–/S‐G2‐Mneurons,Nb:DiI+/Prom1–/G1‐G0.

The mRNA extraction, transcription, amplification and library preparation wereperformedusingcommerciallyavailablekits.For the transcription incDNApoly‐dTprimerandtemplateswitchingoligoswereused.FortheseexactandfurtherkitsseeFlorioetal.,2015.

To identify the genes in the transcriptome the reads were aligned to the mousereferencegenome(mm10)andthehumangenome(hg19).Thedata isexpressedinfragmentsperkilobasepermillionreads(FPKM)withathresholdbiggerthan1.

Methods 3

27

3.6 Histology

3.6.1 Tissuepreparation

For the histological analysis of the brain, the tissue was fixated with differentprotocols.

Toanalyzetheembryosatdifferenttimepointsthemotherwasanesthetizedwith2‐3%Isofluran,theembryoswereexposedanddecapitated.Themotherwassacrificedbycervicaldislocation.Thewholeembryonicheadwasimmersionfixedin4%ofPFAovernight.

The animals analyzed at time points older than P0 were sacrificed by cardiacperfusion. Here animalswere anesthetizedwith intra peritoneal (i.p.) injections ofKetamin (100 mg/ml) and Xylazin (20 mg/ml) solution. The thorax was carefullyopened and the heartwas exposed. PBS and4%PFAwerewashedby a 25 gaugecannula through the left ventricle to clear themouse body of blood and fixate thetissue. 7 ml or 50 ml solution was used for neonatal or juvenile/adult animalsrespectively. The brain was carefully removed and immersion fixated in 4% PFAovernight.

For cryoprotection the tissue was stored in 15 % sucrose in phosphate bufferedsaline(PBS)untilitsank,thereafterthesampleswerestoredin30%sucroseinPBSforatleast24huntilthetissuesankagain.

ForHCNstaining15%saturatedpicrinicacidwasadded to4%PFAandPBSwasreplaced by phosphate buffer (PB) (see materials). The cryoprotection wasperformedwith10%and30%sucroseinPB,untilthebrainsstoppedfloating.

Allbrainswerepromptlyfrozeninoptimalcuttingtemperaturecompound(OCT)inpeel‐awayembeddingmoldsandstoredat‐80°Cuntilfurthertreatment.

ThesampleswereslicedatacryostatCM3050S(Leica)in14µmthicksectionsandsubsequentlystoredat‐80°Cuntilfurtherstaining.

3.6.2 FluorescentNisslStain

ForNeuroTrace®staining thesectionswere treatedaccording to themanufacturesmanual.NeuroTrace®530⁄615RedFluorescentNisslStain(MolecularProbes)wasdiluted 1:25 and NeuroTrace® 500⁄525 Green Fluorescent Nissl Stain (MolecularProbes) 1:600 in PBS. After the staining procedure the sections were mounted inDAPIFluormount‐G(SouthernBiotech).

3 Methods

28

3.6.3 Immunohistochemistry

Forimmunohistochemistrydifferentprotocolswereappliedtoyieldthebestpossiblestaining.

Tostain for cleavedCaspase3,CNPase,Ki67andneuron‐specificclass IIIβ‐tubulin(TuJ1),sectionswerewashedinPBStoremovetheOCT.DetectionofcleavedCaspase3andCNPaseneededanantigenretrieval,thereforthesectionwereincubatedat70°C in10mMsodiumcitratebufferwithpH9 for30min.Afterheating, thesectionswerecooleddowninthebuffer.Unspecificbindingoftheantibodieswasreducedbyblockingwith 5%normal goat serumand0.2%TritonX‐100 inPBS for 1 hour atroom temperature. The primary antibody was diluted in PBS in differentconcentrations (cleaved Caspase 3: 1:2000; CNPase: 1:1000; Ki67: 1:250; TuJ1:1:100)and100µl/slidewereincubatedovernightat4°Cinahumidchamber.Forthe detection of cleaved Caspase 3 the primary antibody was added for 2 days.Thereafter, the slideswerewashed three times for 15min in PBS on a shaker. Todetect the cleaved Caspase 3 and Ki67 antibodies anti‐rabbit secondary antibodydiluted1:200inPBSwasused.AntibodiesagainstTuJ1andCNPaseweredetectedbyincubationwithanti‐mouseantibody(1:200) inPBS.All secondaryantibodieswereappliedfor2hoursatroomtemperatureinahumidchamberinthedark,followedbythreetimes15minwashingwithPBSonashaker.

To detect GFP in brain sections a primary antibody directly coupledwith anAlexaFluor®488fluorophorewasused.ThisstainingwasperformedasdescribedforKi67andTuJ1,withaprimaryantibodydilutionof1:200.Afterthisincubationthesectionswerewashedandmounted.

Staining for neuronal nucleus antigen (NeuN) required antigen retrieval in 10mMsodium citrate buffer with pH 9 for 30 min at 70 °C. Sections for NeuN and glialfibrillaryacidicprotein(GFAP)stainingwerefurtherblockedby10%horseserum,0.2% bovine serum albumin (BSA) and 1% TritonX‐100 in PBS, followed by theprimary antibody diluted (NeuN: 1:250; GFAP: 1:500) in carrier solution (seematerial section) and incubated overnight at 4 °C in a humid chamber. After fourtimes5minwashinginPBS,thesecondaryantibody,anti‐mousediluted1:500,wasappliedin0.5%TritonX‐100inPBSfortwohoursatroomtemperatureinahumidchamber in thedark.The sectionswerewashedagain four times5min inPBSandmounted.

TodetecttheHCNsubunitsinbrainsectionstheslideswerewashedthreetimesfor5mininPB,followedbyblockingfor1hourinHCN‐IHCsolution(seematerialsection).Theprimaryantibodywasdiluted1:5 forHCN4inHCN‐IHCsolutionand incubatedovernightat4°Cinahumidchamber.Sectionswerewashedfourtimesfor5minin

Methods 3

29

PBand thesecondaryanti‐ratantibodywasdiluted1:500 inHCN‐IHCsolutionandincubated for 1 hour at room temperature in a humid chamber in the dark. Thesectionswerewashedagainfourtimes5mininPBandmounted.FordoublestainingwithTuJ1antibody theprotocoldescribed forHCNsubunitswasappliedwithTuJ1antibody diluted 1:100 and secondary anti‐mouse antibody 1:200. HCN4 AntibodywasakindgiftfromPDDr.Wachten.

After the staining procedure all sections were mounted in DAPI Fluormount‐G(SouthernBiotech).

3.6.4 Microscopy

Overview images of the total brain or brain sections were performed with theOlympus SZX16 Binocular with Olympus DP7Z camera with the correspondingsoftwarecellSenseEntry1.4.1.OthertotalbrainpicturesweretakenwithUSBDigitalMicroscopeV2.0anditssoftwareMicroscopeCaptureV2.0fromDyntronic.

FurthermoremicroscopicimageswerecapturedwithanAxioImagerfromZeisswithAxioCamMRcorAxioCamMRmcamerasusingtheZen2litesoftware.

Images using a confocal microscope were captured using a Olympus FluoView™FV1000withFluoviewFV1000cameraandFluoview10softwareversion4.0orbyusingaZEISSLSM880withAiryscanwithZenBlacksoftwareversion2.1SP1.

3.6.5 Stereology

StereologicalanalyseswereperformedwiththeStereoInvestigatorsoftwareVersion10.01fromMicroBrightFieldataNikoneclipse90imicroscope.Fortheproliferationandapoptosisstudiesfoursections84µmapartwerecountedfortheSVZ/VZ,whilethesensorimotorcortexwasrepresentedbyeightslices.Thesectionsfortheanalysisof VZ/SVZ were approximately at 2.31 mm from bregma (Franklin and Paxinos,2007).Fortheinvestigationofthesensorimotorcortextheregionsofinterest(ROIs)were at approximately 2.31mmand3.75mm frombregma. For exampleROIs seeFigure 6‐1. The optical disector principle was used to estimate cell densities asdescribedpreviously(Schmalbachetal.,2015).Theparametersfortheanalysiswere:guard space depth 2 μm, base and height of the disector 3600 μm2 and 10 μm,respectively, distance between the optical disectors 60 μm, objective 40 x Plan‐Neofluar®40x/0.75(Zeiss).

In the migratory analysis five sections 56 µm apart were analyzed. TheelectroporatedpatchwasdividedinlayerII/IIIlayerIVtoVIwithintermediatedzone(IZ) and cortical plate as migratory zone and VZ/SVZ region. See Figure 4‐8 forexampleROIs.AllGFPexpressingcellsinthisROIwerecounted.

3 Methods

30

Tocalculatethecellnumberthefollowingequationwasused:

3.7 WesternBlot

Fordetectionofdifferentproteins,WesternBlotanalysiswasperformed.

Proteinswereextractedfromthewholeheadoftheembryos,frompostnatalday0ontheforebrainwascarefullydissected.Forthispurpose,themotherwasanesthetizedwith2‐3%ofIsofluranandtheuterusandembryoswereexposed.Themotherwassacrificed by cervical dislocation. For the adult protein analysis the forebrain wasremoved shortly before sacrificing the animal by cervical dislocation. All tissuesamples were immediately frozen in liquid nitrogen and stored by ‐ 80 °C untilfurthertreatment.

Toinhibitthedegradationoftheproteins,allfollowingstepswereperformedonice.Thesampleswerehomogenizedusinga1.5mlpistiland200–400µlsucrosebuffer(seematerials). The sampleswere centrifugedwith 5000 g for 10min at 4 °C thepelletwasre‐suspendedandcentrifugedasecondtimeunderthesameconditions,toremove cell debris and the nucleus. To extract only the membrane proteins thereunited supernatant was centrifuged at ~ 100,000 g for 45 min at 4 °C. Thesupernatant contains the cytosolic proteins, while the membrane proteins in thepelletarere‐suspendedin50–100µlsucrosebufferdependingonthepelletsize.

Tovalidatetheproteinconcentrationofthesamplesabicinchoninicacidassay(BCA)assay from Thermo Fisher was used in accordance to the manufacturer protocol.Sampleswerediluted1:100and1:50andmeasuredinanEpochreaderfromBioTekanditsaccordingsoftwareGen52.07.

TheseparationoftheproteinsbysizewasperformedwithaNuPAGE®Bis‐Trisgelelectrophoreses(Invitrogen)withapolyacrylamidegradientof4‐12%.

Gelswith 10wellswere used. Approximately 20 µg of total protein perwellwereapplied. The sampleswere treated accordance tomanufacturer protocol andwereloaded in a XCell SureLock™ electrophoreses chamber (Invitrogen). In NuPAGE®MOPSSDSrunningbufferthegelswererunforapproximately30minat150mV.Asmarker6.5µlofSpectra™MulticolorbroadrangeproteinladderfromFermentaswasused.

Methods 3

31

Todetect theproteinof interestviaantibodies theproteinshave tobeblotted toanitrocellulosemembrane.For thispurposeaMini‐Protean®chamber fromBio‐Radcontaining NuPAGE® transfer buffer with 10 % methanol was used. Proteins arenegativelychargedsothemembranehastobedirectedtotheanodeofthechamber.Blotting is performed for 2 hours at 80 V and 30min at 88 V or overnight at 6 Vfollowedby30min85V.

Todetecttheproteinofinterestonthenitrocellulosemembrane,differentantibodieswereused.Todetect the transgene,which iscoupledtohemagglutinin(HA) tag,ananti‐HAantibody(Roche)wasused.ThedifferentHCNsubunitsaredetectedviaanti‐HCN1,anti‐HCN2,anti‐HCN3,andanti‐HCN4(allmouse,neuromab).

Toblockunspecificreactionsoftheantibodiesthemembranewasblockedwith5%milk powder and 0.05 % Tween20 in TBS for 1 hour at room temperature on ashaker.Subsequently6mlofanti‐HA1:500andanti‐HCN1‐41:2000dilutedin1%milkpowderand0.01%Tween20inTBSwasadded.Themembranewasincubatedfor1houratroomtemperatureorovernightat4°Conashaker.Toremoveunboundantibodiesthemembranewaswashedfourtimes5minin0.05%Tween20inTBSatroom temperature. The secondary antibody is conjugated with horse radishperoxidase(HRP)andwasdiluted1:500in1%milkpowderand0.01%Tween20inTBS.Thenitrocellulosemembranewasincubatedin6mlofthesecondaryantibodysolutionfor1houratroomtemperature.Afterwardsthemembranewaswashedfourtimes5minin0.05%Tween20inTBS.Todetectthespecificallyboundantibodies2mlofLuminata™CrescendoWesternHRPSubstrate(Millipore)wasaddedfor5min.TheproteinbandsaredetectedintheAdvancedFluorescent Imager fromIntasandanalyzedwiththeaccordingsoftwareChemoStarImager.

3.8 Inuteroelectroporation

For the in utero electroporation (IUE) plasmids were prepared with an EndoFreePlasmid Maxi Kit from Quiagen according to manufacturer protocol. The usedconstructsintheproliferationandmigrationexperimentswerepCAGig‐GFP,pCAGig‐HCN‐wt‐GFP and pCAGig‐HCN‐DN‐GFP. Constructs were cloned by Kathrin Sauterusing thepCAGig‐GFPvector (addgene) cutwithEcoRVand ligatedwithHCN4‐WT(wtHCN4‐pBUD_EGFP; AseI and XbaI) or HCN‐DN (HCN4‐AYApcDNA1;HindIII/Xba1).Seesupplement6.2forvectorcharts.

TheIUEforthemigrationexperimentswereperformedinC57Bl/6JdamsbredwithmaleC57Bl/6JatE15.

3 Methods

32

FortheIUEthedamwasinjectedi.p.with0.05mg/kgofbodyweightbuprenorphine0.5hoursprecedentoftheprocedure.Duringthesurgerythedamwasanesthetizedwith1.5‐2%Isofluran(1000ml/mininair).Thebodytemperatureiscontrolledbyarectalprobe.Toavoidthedehydrationoftheeye, themouseistreatedwithVidisic.After shaving and disinfecting the abdomen with Betaisodona, the uterus wasexposedbysmallincisionsoftheskinandperitoneum.1‐2µlDNA(0.5‐5µg/µl)wereinjectedintotheventricleoftheembryo,byfollowingthemidlineoftheembryoscalpandperforatingtheuterus,skinandbonerightnexttoitwithacapillary.Capillarieswere pulled with a Sutter Puller Model P‐1000 from borosilicate capillaries withfilament(GB100TF‐10,0.78x1.00x100mm;ScienceProducts)aimingforalongandthin tip, which is shortened right before the injection, to sharpen the edges of theglass.ForvisibilityoftheDNAsolutionintheventricle0.05%FastGreenwasadded.Five pulses of 35 V were applied by the 5 mm tweezertrodes (BTX, HarvardApparatus) targeting the cortex of the E15 embryos. After carefully suturing theperitoneumwithanabsorbablesutureandtheabdominalskinwithasilksuture,themotherwasinjectedwith5mg/kgbodyweightofCaprofen,alonglastingpainkillerand anti‐inflammatory. Thewoundwas disinfectedwith Betaisodona and the damwas placed back in her home cage and monitored until total recovery from theanesthesia.

TheembryoswereimmunohistochemicallyanalyzedatE19.

3.9 CellCulture

AllcellcultureexperimentswereperformedbyDr.SabineVayfromtheDepartmentforNeurologyattheUniversityhospitalCologne.Forprimaryneuralstemcellculture(NSC) fetal rat cortices were derived from embryonic day 13.5 as describedpreviously(Ruegeretal.,2010).ToDMEM/F‐12medium(Gibco),supplementedwithL‐glutamine,N2‐supplement, Penicillin/Streptomycin and sodiumpyruvate (Gibco),humanrecombinantfibroblastgrowthfactor(humanFGF2;10ng/ml,Invitrogen)wasadded. After two passages, 2 x 104 cells were re‐plated in chamber slides asmonolayers in thepresenceofFGF2for furtherexperimentalprocedures.Toassessthe proliferation potential, NSC were treated with the Ih blocker ZD7288 (Sigma‐Aldrich)in10µMand30µMconcentrationsandRNA‐extractionorBrdU‐assaywasperformed after 18 hours.Differentiationwas initiatedbywithdrawal of FGF2 andZD7288 (30 µM) was added every 24 hours. After 3, 5, 7, and 10 days ofdifferentiation,NSCwerefixedandimmunocytochemicallystained.

Methods 3

33

3.9.1 Invitroproliferationassays

For5‐Bromo‐2’‐deoxyuridine(BrdU)experiments10µMBrdU(Fluka)wasaddedfor6 hours to the cell culture medium before fixation with 4 % PFA for 10 min. Asantigenretrieval2MHClfor30minwasused.Anti‐BrdUantibodywasdiluted1:100.Pursuingwiththeprotocoldescribedin3.9.2.

3.9.2 Immunocytochemistry

To analyze the differentiation in cell culture Sox2 was used as neuronal stem cellmarker(1:100),TuJ1(1:100),asmarkerforearlyneurons,GFAP(1:1000)indicatingastrocytes and CNPase (1:5000) to mark oligodendrocytic precursor cells,additionally cells were stained with HCN4 antibody (1:5). For theimmunocytochemicalstainingcellswerewashedwithPBSandfixatedwith4%PFAfor 10 min. After washing three times for 5 min each with PBS, cells werepermeabilized and blocked in 0.1 % TritonX‐100 and 10 % goat serum in PBS.Primary antibody was applied overnight in 3 % goat serum at 4 °C in respectiveconcentrations.ThricerepeatedwashingwithPBS for5mineachwere followedbysecondary antibody incubation (1 hour at room temperature) with fluoresceine‐labeledanti‐mouseimmunoglobulinG(IgG)oranti‐rabbitIgG(1:200,Invitrogen)in3% goat serum in PBS. Cells were additionally counterstained with Hoechst 33342(Invitrogen) diluted 1:500 and incubated for 5 min. Slides are mounted withFluoromount(SouthernBiotech).Allmicroscopicinvestigationswereperformedwithan inverted fluorescent phase‐contrast microscope (Keyence BZ‐9000E). Cellnumbersineachexperimentwereanalyzedby10imagesperconditionandstaining.

3.9.3 Transwellmigrationassay

Formigratory analysis a transwellmigration assaywas performed using a Boydenchamber system (CytoSelectTM 24‐Well Cell Migration Assay, pore size 8 µm, CellBiolabs,Inc.,SanDiego,USA).Neuralstemcellswereseededinthetopoftheinsert(4x 104 cells)with orwithout ZD7288. After 24 hours, the non‐migratory cellswereremovedandmigratorycellswerestainedwiththecellstainsolutionfor10minutes.Cells were gently washed and air dried. Then, stained cells were dissolved by theExtractionsolutionandopticaldensity(OD)wasmeasuredat560nminamicroplatereader(FLUOstarOmega,BMGLabtech)

3.9.4 ViabilityAssays

LDH release, in the medium of the cultured cells was measured using the LDHCytotoxicity Assay Kit (Thermo Scientific). Briefly, themedium (50 μl)wasmixedwith50μlofreactionmixture.After30min,50μlofstopsolutionwasaddedandthe

3 Methods

34

absorbancewasreadat490nmand680nm.TodeterminetheLDHrelease,the680nmabsorbancevaluewassubtractedfromthe490nmabsorbance.

The LIVE/DEAD® Cell‐Mediated Cytotoxicity Kit (life technologies) was used toanalyze the number of apoptotic cells. NSC were stained, with propidium iodide(1:500) as non‐permeable dye and simultaneously stained with Hoechst 33342(1:500).After5minutesat37°C,cellswerecountedusing the inverted fluorescentphase‐contrast microscope (Keyence BZ‐9000E). Cell numbers in each experimentwereanalyzedby10imagesperconditionandstaining.

3.9.5 Electrophysiology

Neuronalstemcellsweregrownonglasscoverslips,whichweretransferredintothebath solution shortly before the start of the recordings. Pipettes with filament(GB150F‐10,0.86x150x100mm;ScienceProducts)werepulledwithaSutterPullerModel P‐1000 with a resistance between 2‐6 Ω and were filled with electrodesolution.ThemeasurementswereperformedwithanEPC‐9amplifierformHEKAandthePulse+PulseFitv8.79softwarev8.79.Allmeasuredcellshadanuncompensatedseriesresistancefrom<15MΩ.

Results 4

35

4 Results

Tothebestofmyknowledge,thisisthefirststudytorevealaroleofHCNchannelsinearly embryonic development. Due to the novelty of this subject, the endogenousHCN‐subtype expression in the embryonic forebrain first had to be establishedtogether with a mouse line with an early embryonic expression of a forebrain‐restricted dominant‐negative HCN subunit. This dominant‐negative approach wasspecificallydesigned to functionally ablate the Ih in a subunit‐unspecificmanner asearly as during embryonic brain development. Using MRI measurements, themorphological alterations were characterized and quantified and showed apronounced microcephaly phenotype. This phenotype was further investigated invivo and in vitro, where the different developmental processes of the corticaldevelopmentweredissectedandanalyzed,namely theproliferation,differentiation,and migration of neural stem cells, progenitor cells, and young neurons. Toinvestigate themechanism(s) underlying themicrocephaly, patch‐clamp recordingsof primary cortical stem cellswere conducted and showed the electrophysiologicalprofileoftheanalyzedcells.

4.1 EndogenousHCNexpressionduringembryonicdevelopment

TheexpressionofHCNsubtypesinthepostnatalbrainofrodentshasbeenpreviouslyanalyzed (see1.3.2).However,nodatawereavailablewith respect toHCN‐subtypeexpressionintheembryonicbrain.ToinvestigatetheimpactofIhonprenatalcorticaldevelopment, theexpressionpatternof thedifferentHCNsubtypeswasanalyzedatthemRNAandproteinlevels.

4 Results

36

To first assess the mRNA expression of the different subtypes a qPCR assay wasperformed to analyze the total mRNA of the mouse telencephalon at variousdevelopmental timepoints in the embryonic brain (E13, 15, and18) andpostnataltissue (P7 and P35) (data reanalyzed from Schlusche, 2011). HCN1 and HCN2expressionincreasedduringdevelopment,reachingthehighestlevelsatadultstagesandrepresentingthemainHCNsubtypesintheadultforebrain(Figure4‐1A;upperpanel),asreportedinpreviousliterature(see1.3.2).Notably,andincontrasttoHCN1

Figure4‐1ExpressionofHCNsubtypesduringdevelopment

A,PooledmRNAfromthetelencephalonwasanalyzedinaqPCRassayatdifferentdevelopmentaltimepoints, namely at embryonic days (E) 13, 15, 18, andpostnatal days (P) 7 and35.HCN1andHCN2expression levels increase throughout development, reaching their highest levels at adult stages(upperpanel),whileHCN3andHCN4expressionpeaksatE15(lowerpanel).DatawerenormalizedtoGAPDHandnormalizedtothevalueatE13.B,RNAsequencingofapicalradialglialcells(aRG),basalradialgliacells(bRG),basalintermediateprogenitors(bIP)andneurons(N)oftheE14.5mousecortex(upperpanel)andaRGandbRGaswellasofneuronsinthehumanembryo13weekspostconception(wpc13; lowerpanel)wasperformed. Inall cell types,HCN3exhibits thehighestexpression(FPKM:fragments per kilobase permillion reads, dotted line: noise threshold; performed by Dr. Florio).C,WesternblotanalysisinE15whole‐headmembraneproteinextractions,showsHCN3andHCN4asthemain subtypes in the embryonal brain.D, Immunohistochemistry of E13murine cortex reveals co‐expressionofHCN4andTuJ1,anearlyneuronalmarker.Scalebars,50µm

Results 4

37

and HCN2, the relative expression of HCN3 and HCN4 peaked at E15 andsubsequentlydecreasedagain,reachinglowlevelsintheadulttelencephalon(Figure4‐1A;lowerpanel).ThisexpressionpatternsuggestsanalterationintheexpressionofthemajorHCNsubtypesduringdevelopment.ThedataissupportedbyanmRNAnext‐generation sequencing study, performed and reanalyzed by Dr. Marta Floriofrom the laboratoryofProf.Huttner. In thedevelopingmurine cortexdifferent celltypesweredissected,namelyaRG,bRG,bIPandneurons(Florioetal.,2015).Inthisexperiment,HCN3wasthedominatingsubtypeinallcelltypesanalyzed(Figure4‐1B; upper panel). In the human embryo, 13weekspost conception (wpc), aRG, bRGandneuronsweredistinguished(Florioetal.,2015). Inaccordancewiththerodentbrain,HCN3was thesubtypewith thehighestexpression, increasing fromapical tobasalRGuntilitreachedthehighestexpressionlevelinneurons(Figure4‐1B;lowerpanel).

Attheproteinlevel,thesedatawereconfirmedwithrespecttothefindingthatHCN3and HCN4were themost highly expressed subtypes. In western blot analysis, themembrane proteins from the whole head of E15 embryos were extracted andcomparedtoadultpreparations,detectingonlyHCN3andHCN4(Figure4‐1C).ThesedataconfirmthechangeindominantHCNsubtypes,whichwasalsosuggestedbythemRNAlevelsofthetelencephalon.TolocalizetheHCNchannelatthisearlytimepointin the evolving cortical structures, immunofluorescence experiments wereperformed.TheHCN4subtypewasdetectedattheoutersurfaceofthecorticalplateinthemurineforebrainatE13(Figure4‐1D;leftimage).Thisexpressionco‐localizedwithβtubulintypeIII(furthercalledTuJ1)expression,amarkerforyoungneurons(Figure 4‐1 D; center image). The HCN3 antibody did not yield specific stainingresults.

In summary, these experiments show the expression of HCN channels in theembryonicbrain,withaswitchbetween thedominantsubtypes fromHCN3and/orHCN4 in earlyprenataldevelopment toHCN1andHCN2 in thepostnatal andadultbrain.

4.2 GenerationofanembryonicHCN‐DNmouseline

To investigate the role of HCN channels and of Ih in early prenatal corticaldevelopment, theEMX1‐promotermouse linewasused.EMX1 isexpressedfromasearlyasE9.5inaforebrain‐specificmanner(Iwasatoetal.,2000;Gorskietal.,2002).Bythisforebrainrestrictionotherunwantedphenotypessuchasembryoniclethalityofthecardiacexpressionwasavoided(see1.3.4).

4 Results

38

TovalidatetheknownexpressionpatternoftheEMX1‐Cremouseline(Iwasatoetal.,2000), thesemice were bred with the Ai9 line, expressing a floxed stop sequencefollowed by the CAG promoter (cytomegalovirus enhancer combined with chickenbeta actinpromoter) and the tandemdimer (td)Tomato red reporter gene (Figure4‐2A).Creexpressionresultsinthetranscriptionofthereportergenebyexcisionofthe prepending stop sequence in theRosa26 locus. The forebrain restrictionof theEMX1 promoter was demonstrated in sagittal sections of the mouse brain,

Figure4‐2EMX1drivenmouseline

A,TovalidatetheEMX1‐promoterexpression,theEMX1‐CrelinewascrossedwiththeAi9mouseline,expressingthereporter tandemdimer(td)Tomatounder thecontrolof thehighly transcribedCAG‐promoter(cytomegalovirusenhancercombinedwithchickenbetaactinpromoter)intheRosa26locus.B, Images show fluorescent Nissl staining (green) in sagittal brain sections of the EMX‐Ai9 mouse(tdTomato in red); the overlay is depicted in yellow. The expressionwas forebrain restricted, onlyvisibleindentategyrus(upperleft),CA3(upperright)andcortex(lowerleft)butabsentincerebellum(lower right). C, In the experimental mouse line EMX‐HCN‐DN, expression is driven by the EMX1‐promoter resulting in the expressionof the cre recombinase,which excise the stop sequence in theRosa26 locus leading to the expression of the tetracycline transactivator (tTA). In the absence ofdoxycycline, the tTA binds to the responsive element (Ptet), a bidirectional promoter, inducing theexpression of the human dominant negative HCN subunit (hHCN‐DN) and EGFP. D, The HCN‐DNsubtypeexhibited theG480Smutation in theGYGmotifof theporeregion.E,Westernblotanalysisshowed the expression of the transgene fromE10.5 onward, detectedwith an antibody against thehemagglutinin‐tagcoupledtothetransgene.Scalebars,100µm

Results 4

39

counterstained with fluorescent Nissl in green (Figure 4‐2 B). TdTomato wasdetected in the granular cells of the dentate gyrus, excluding the subgranular celllayer. In the cornus ammonis region 1 (CA1) of the hippocampus, the somata ofgranulecellsexpressed tdTomatoaswell as themossy fibers inCA3 (Figure4‐2B;upper panel). In the cortex, expression of tdTomato was mainly detected in theneuropil (Figure 4‐2 B; lower left). The localization of the tdTomato suggests apreferenceof thereporter for theprocessesof theexpressingcells.Thecerebellumshowednoexpression,supportingtheformerlydescribedforebrainrestrictionofthepromoterline(Figure4‐2B)(Iwasatoetal.,2000).

ToevaluatethefunctionalroleofHCNchannelsduringdevelopment,ourlabcreateda triple transgenic mouse line in which the spatial and temporal expressions of adominant‐negative HCN subunit were controlled (Figure 4‐2 C). To achieve anembryonic expression of the transgene, the EMX1‐Cre promoter line was crossedwith a linker line carrying a floxed stop sequence followed by the tetracyclinetransactivator(tTA)intheRosa26locus.Duetotheactivationofthecrerecombinase,thestopsequence is removed, resulting in theexpressionof the tTA,which in turnbindstoitsresponsiveelement(Ptet)intheabsenceofdoxycycline.ThetTAandPtetresemble the Tet‐off system, which gives additional temporal control of theexpression pattern by the application of doxycycline. The administration ofdoxycyclinethusleadstotheinactivationofthetransgeneexpression.Intheabsenceofdoxycycline,Ptet,asabi‐directionalpromoter,isactiveandleadstotheexpressionof both the human dominant‐negative HCN subunit (hHCN‐DN) carrying ahemagglutinintag(HA)andthereporterEGFP.TheHCN‐DNisapreviouslydescribedHCN4subunitwiththeG480SmutationintheGYGmotifofthechannelporeresultinginanSYGmotif.Themutation leads toanon‐conducting channel and to a subunit‐independent ablation of Ih (Figure 4‐2D; comparewith Figure1‐2) (Sandke, 2006;Merseburg,2016).

Inwesternblotanalysis,thetransgeneexpressionwasshowntobepresentasearlyas E10.5 (Figure 4‐2 E). Due to technical limitations,membrane protein from E9.5embryos could not be extracted in amounts high enough for analysis. Therefore,transgene expression at this early time point could not be proven. The HCN‐DNsubunitisdetectedviaahemagglutinintag(HA‐tag)‐bindingantibody,todistinguishthetransgenicsubunitfromendogenousHCNsubtypes.

The spatial restriction and the early expression of the EMX1 ‐ promoter incombination with the subunit‐independent suppression of Ih resulted in the firstreported approach to the analysis of HCN‐channel activity at early embryoniccerebraldevelopment.Thus,itwasthefirstinvivoinvestigationoftheconsequencesofalternatingHCN‐channelfunctioninneuralprogenitorcellsandearlyneuronsinprenataldevelopmentalprocessesofthemurinecortex.

4 Results

40

4.3 MorphologicalcharacterizationofEMX‐HCN‐DNmice

ToanalyzetheroleofHCNchannelsinembryonicdevelopment,thebrainstructureofHCN‐DN mice were first characterized. EMX‐HCN‐DN mice showed marked andsevere morphological abnormalities as early as at P0. The pups were viable untilweaning, but then showed increased lethality at approximately threeweeksof age.This early lethality did not allow an analysis of adult but only of juvenile (P21)animalstobeperformed.

AsearlyasatP0,thebodyweightwasdecreasedinmutantEMX‐HCN‐DNmice(mut)compared to control animals (con) or mutant mice, which had been fed withdoxycyclineuntilP0(mut+dox)(Figure4‐3A).AtP21,thebodysizeofmutantmice

Figure4‐3MorphologicalabnormalitiesinEMX‐HCN‐DNanimals

A,Atpostnatalday(P)0,thebodyweightisdecreasedinmutantmice(mut)comparedtocontrol(con)ortransgenicanimalstreatedwithdoxycyclineuntilP0(mut+dox).B,ThedecreaseinbodysizeandweightismorepronouncedinmutantanimalsatP21comparedtocontrolormut+dox.C,Neonatalmutant mice show a decrease in cortex size at macroscopic and microscopic images. The coronalsectionsstainedwith red fluorescentNissldye reveal theabsenceofahippocampal structure.D,AtP21, thedecrease incortical structureswasvisible in the totalbrain (topview).Thereducedcortexand the absent hippocampusare clearly visible in the overview image of the brain and in theNisslstained coronal section. E, The phenotype was rescued by the application of doxycycline until P0,suppressing the transgene expression until the day of birth, resulting in morphologically normalbrains.Scalebars,Cleft1mm,DandEleft2mmin;allNisslstainingimages500µm*p<0.05,**p<0.01, ***p<0.001,****p<0.0001;Dataarepresentedasmean±S.E.M;one‐wayANOVAwithpost‐hocTukey´stest;A,con:n=14,mut:n=8,mutdox:n=8;B,con:n=15,mut:n=10,mutdox:n=8.

Results 4

41

was drastically reduced in comparison to control mice (Figure 4‐3 B; left). Thissmallerbody sizeof themutantmicewasparalleledbya severe reduction inbodyweight by approximately 50% compared to control animals (Figure 4‐3 B; right).WhenadministeringdoxycyclineuntilP0inmutant+doxmice,thispronouncedbodyweight reduction was prevented (Figure 4‐3). There was no detectable differencebetweenthecontrolanimalswithandwithoutapplicationofdoxycycline.

Inadditiontotheoveralldecreaseinbodymassand‐size,acloserinvestigationofthebrain of neonatal EMX‐HCN‐DN mice revealed a dramatic decrease in the size ofcortical structures (Figure 4‐3 C; left). The reduced cortex does not cover thediencephalonormesencephalonas seen in thecontrol animals. In fluorescentNisslstaining (red) of coronal sections the smaller cortex and the absence of thehippocampal structures were pronounced (Figure 4‐3 C; right). This phenotypepersisted until P21, leaving the diencephalon and the lateral and third ventriclepartiallyuncovered(Figure4‐3D).Intheneonatal,andinthejuvenileanimals,GFP‐positivetransgeniccellsaccumulatedatthebasalcortex,liningtheapicalsideofthelateral ventricles (Figure 4‐3 C and D). In mice treated with doxycycline duringprenatal development, suppressing the expression of the HCN‐DN subunit, nomorphological abnormalities were observed (Figure 4‐3 E). The GFP‐positive cellswere found in thesubgranularzoneof thedentategyrusandtheventricular lining,theareasofpostnatalneurogenesis(Figure4‐3E,lowerright).

To quantify the severe loss of forebrain structures, MRI measurements wereperformed. Preliminary MRI data were published in Schlusche, 2011, but wereextendedandfinalizedinthecourseofthisdoctoralthesis.

4 Results

42

The reduction was quantified by three‐dimensional (3D) volume rendering of thebrainandventriclesinneonatalandjuvenileanimals(Figure4‐4AandB)(Schlusche,2011). The brain volume of neonatal mutant mice was significantly decreased byapproximately25%comparedtocontrolandmutant+doxanimals(Figure4‐4C).Incontrast,theventriclevolumewasnotchangedbetweenthegroups(Figure4‐4D).Injuvenile mutant animals, the brain volume was even further reduced byapproximately50%comparedtocontrolandmutant+doxanimals(Figure4‐4E).Additionally,therelativeventriclesizeofmutantanimalswasincreasedbynearly50%incomparisontocontrolandmutant+dox(Figure4‐4F).

Insummary, inP0andP21EMX‐HCN‐DNmice,severemorphologicalabnormalitiesweredetectedalongwithdecreasedcortexsizeandtheabsenceofthehippocampalstructures.UponsuppressionofHCN‐DNexpressionuntilbirth, thisphenotypewasrescued, suggesting an important functional role for HCN channels in prenataldevelopmentalprocessesoftheforebrainstructures.

Figure4‐4QuantificationofmorphologicalabnormalitiesinEMX‐HCN‐DNanimals

A,B, Representative images ofMRI data and the 3D rendering of the control brain volume (upperpanel)andtheventriclevolume(lowerpanel)wereconductedatP0(A)andP21(B).DepictedaretheMRI images with two different perspectives on the 3D model. C, The brain volume in cm3 wassignificantly decreased in mutant (mut) compared to control (con), and was rescued by theadministrationofdox(mut+dox).D,AtP0,theventriclevolumeinpercentofthetotalbrainvolumewasnot affectedby Ih ablation.E, AtP21, thebrain volumewas severelydecreased inmut animalscomparedwith conandmut+dox.F, Theventricle volume in juvenilemutmicewas increased;nodifferencebetweentheconandmut+doxwasdetectable.* p < 0.05, ** p < 0.01, *** p < 0.001, **** p < 0.0001;Data is presented asmean± S.E.M; one‐wayANOVA with post‐hoc Tukey´s test; C, and D, brain volume: con: n=14, mut: n=8, mut dox: n=8;ventriclevolume:con:n=12,mut:n=4,mutdox:n=7;E, andF,brainvolume:con:n=16,mut:n=10,mutdox:n=8;ventriclevolume:con:n=15,mut:n=10,mutdox:n=8.

Results 4

43

4.4 AnalysisofdevelopmentalprocessesintheEMX‐HCN‐DNmouse

TodissecttheroleofHCNinprenataldevelopmentinmoredetailandtoidentifythemechanisms affected by the ablation of Ih causing the reduction in brain size, theproliferation,differentiation,andmigrationofneuronalprogenitorcellsandcorticalneuronswereinvestigatedinEMX‐HCN‐DNmice.Furthermore,apoptosis,apotentialmechanismleadingtoreducedbrainsize,wasanalyzed.

4.4.1 ProliferationinEMX‐HCN‐DNmice

To investigate the overall proliferation in the forebrain of EMX‐HCN‐DN animals,immunostaining with the Ki67 antibody was performed at P0. Ki67 is a nuclearprotein expressed throughout the cell cycle from G1 to M phase, markingproliferatingcells(Gerdesetal.,1991).

Figure4‐5ProliferationinEMX‐HCN‐DNmiceatP0

A, Example images showcoronal sections of control (con) andmutant (mut)EMX‐HCN‐DNmicewithKi67 staining (red) in the sensorimotor cortex and ventricular zone/ subventricular zone (VZ/SVZ).SectionswerecounterstainedwithDAPI (blue),GFP(green)expressionmarkedthe transgeniccells inmutmice.B,Stereologicalquantificationofsensorimotorcortexshowednosignificantdifference(upperpanel), but in the proliferation zone VZ/SVZ mitotic cells were overall significantly reduced. Datadepictedindensity(D)in[cells/mm3].Scalebars,100µm;*p<0.05,**p<0.01,***p<0.001,****p<0.0001;Dataispresentedasmean±S.E.M;unpairedt‐test,cortex:con:n=6,mut:n=5;VZ/SVZ:con:n=4,mut:n=4.

4 Results

44

The proliferationwas analyzed in the sensorimotor cortex and theVZ/SVZ (Figure4‐5 A). Coronal sections were stained with Ki67 in red, counterstained with DAPI(blue), and the transgenic cells expressed GFP (green). In the control animalsproliferationwasdetectedinbothregionsofinterestandinthedentategyrusofthehippocampus (Figure4‐5A; left panel). ElevatedKi67 expressionwas found in theVZ/SVZandinthedentategyrus,regionsofproliferationatthispostnataltimepoint,whileproliferatingcellswerealsodetectableinthecortex,butinsmallernumbers.Inthemutantsections,Ki67‐positivecellsweredetectedinallstructuresanalyzed.TheVZ/SVZwasclearlymarkedwithKi67staining.TransgeniccellsexpressingGFPweredetected in the remaining cortex, with accumulations proximal to the lateralventricles (Figure 4‐5 A; right panel). Stereological quantification revealed nosignificantdifferenceinthenumberofproliferatingcells inthesensorimotorcortexbetweencontrolandmutantEMX‐HCN‐DNmice,butadecreaseinmitoticallyactivecellsintheVZ/SVZ(Figure4‐5B).ThereductioninbrainsizecouldthusbeduetothelocaldecreaseinmitoticallyactivecellsintheproliferationzoneVZ/SVZ.

ToverifywhetherthesizereductioninforebrainstructuresseeninEX‐HCN‐DNmicewas due to decreased stem and progenitor cell proliferation, the NEXCre‐HCN‐DNmouselinewasinvestigated.NEXCreisexpressedfromE10.5onwardinpost‐mitoticneurons,mainlyinforebrainstructures(Goebbelsetal.,2006)and,thus,resembledthe expression pattern of EMX1, excluding proliferating precursor cells. Thisapproach allowed the process of proliferation to be dissected from other prenataldevelopmental processes, such as differentiation and migration, without analyzingtheembryosandtherebysacrificingthemother.

Results 4

45

To generate NEXCre‐HCN‐DN mice, a genetic approach similar to that earlieremployedforEMX‐HCN‐DNmicewasused(Figure4‐6A).Theexpressionofthecrerecombinasewas driven by theNex promoter using the above‐described temporaland spatial expression pattern. Cre recombinase‐excised the floxed stop sequencepreceding the tTA, located in the Rosa26 locus. The tTA bound in the absence ofdoxycycline to the bi‐directional Ptet, allowing the HCN‐DN subunit and GFP to beexpressed.

Figure4‐6MorphologicalnormalbrainsofNEXCre‐HCN‐DNmice

A,CrerecombinasewasexpressedundertheNEXpromoter,leadingtotheexcisionofthefloxedstopsequence intheRosa26locus, thusallowingthetetracyclinetransactivator(tTA)tobeexpressed. Intheabsenceofdoxycycline,thedominant‐negativeHCN‐DNsubtypeandEGFPareexpressedthroughthebidirectionalresponsiveelement(Ptet)oftTA.B,TheoverallmorphologyoftheadultNEXCre‐HCN‐DNmicedidnotdifferfromcontrols.C,CoronalsectionsstainedwithfluorescentNissl(red)shownoalterationinthegeneralstructureofthecortex,hippocampus(CA1anddentategyrus)ortheventricle.Scalebars,inB,2mmandinC,100µm.

4 Results

46

In contrast to the microcephaly phenotype of EMX‐HCN‐DN animals, the NEXCre‐HCN‐DNmutantanimalsexpressingtheHCN‐DNsubunitfromanearlyprenataltimepointonwardshowednogrossmorphologicalalterationsinthebrainattheageofsixweeks.Thevisiblecorticalstructurescoveredthediencephalonandmesencephalon,leavingonlythecolliculiinferioresandsuperioresuncovered(Figure4‐6B).Coronalsectionsstainedwith fluorescentNissl confirmed intactmorphologiesof thecortex,CA1, anddentate gyrusof thehippocampus andventricles (Figure4‐6C).GFPwasdetected,asreporterfortransgeniccells,intheneuropilofthecortexandCA1andafew cell somata in both regions. In the dentate gyrus, GFP‐expressing cells weredetected in the granular cells and their dendrites, excluding the subgranular celllayer,theregionofneuronalprecursorcellspotentiallyundergoingmitosis.AnalysisofthesectionsshowingthelateralventriclesrevealedGFPinthelowercorticallayersand fiberbundlesof the striatumandcorpus callosum,butnot in theVZ/SVZ.Thisobservation is in line with the expression pattern of the NEX promoter, which isactiveinpost‐mitoticneurons,butnotinproliferatingcells(Goebbelsetal.,2006).

Insummary,observationsofadecreasedproliferationinP0EMX‐HCN‐DNmiceintheVZ/SVZ, together with the morphologically normal forebrain of NEXCre‐HCN‐DNadultmice, suggest an impaired proliferation of neural progenitor cells, due to theablatedIhinearlyprenataldevelopmentofthemurineforebrain.

4.4.2 DifferentiationintheEMX‐HCN‐DNmouseline

The differentiation of neural progenitor cells into neurons, astrocytes, andoligodendrocytes is another important mechanism of cortical development. ToanalyzethedifferentiationofHCN‐DN‐expressingcells intheforebrain,brainsofP0animalsweresectionedandstainedformarkersofeachcelltype.TuJ1,wasusedasamarker of early neurons, astrocytes were detected with an antibody against glialfibrillary acidic protein (GFAP) and 2',3'‐cyclic‐nucleotide 3'‐phosphodiesterase(CNPAse)wasusedforthestainingofoligodendrocytes(Engetal.,1971;Verrieretal.,2013).

Results 4

47

ThehippocampuswasusedasalandmarkintheP0brainsofcontrolanimals,andthesame brain regions were matched in the EMX‐HCN‐DNmice (Figure 4‐7 A). TuJ1,indicating early neurons, was expressed throughout the brain in control and inmutant EMX‐HCN‐DNmice.A partial overlaywith the transgenic, Ih ‐ ablated, GFP‐positivecellswasdetected,andisshowninyellow(Figure4‐7B).GFAP,amarkerforastrocytes(Figure4‐7C)andCNPase,expressedinoligodendrocytes(Figure4‐7D)wasnotdetectedinthecontrolanimals(Figure4‐7CandD,upperpanel).Incontrast,bothglialmarkerswereexpressedinEMX‐HCN‐DNmiceatP0(Figure4‐7CandD).GFAPandCNPasewereexpressedonthesurfaceoftheremainingcorticalstructure,suggesting the formation of a glial scar‐like structure. In control animals, theexistence of the neuronal marker and the absence of both glial markers are in

Figure4‐7DifferentiationinP0EMX‐HCN‐DNmice

A,Thecartoonofacoronalsectionshowsthesectionalplaneofthecontrolandmutantanalyzedbrainslides,from(FranklinandPaxinos,2007).B,TuJ1(red)asamarkerforearlyneuronswasexpressedincontrol as well as EMX‐HCN‐DN mutant mice, showing a partial overlay with the GFP expressingtransgeniccells(yellow).C,AstrocyteswerestainedwithGFAP(red),amarkerwasonlyexpressedinP0EMX‐HCN‐DNanimals,butnotincontrols.C,CNPaseasamarkerforoligodendrocytes(red)isonlydetectable inmutant but not in controlmice. Both glial markers are present on the surface of theremainingcorticalstructures.AllcoronalsectionswerecounterstainedwithDAPI(blue).Scalebars,100µm.

4 Results

48

agreementwiththetemporalpatternofcorticaldevelopment,withtheneurogenesisatearliertimeperiods,followedbythegliogenesisaroundbirth(see1.1).

TheexpressionofglialmarkersinP0mutantssuggestsanalterationinthetemporalsequenceofthedifferentiationpatternintheIh‐deficientforebrainsofEMX‐HCN‐DNmice.

4.4.3 MigrationofcellswithIhablation

ItwasnotpossibletodirectlyinvestigatethemigrationofneuronsintheEMX‐HCN‐DNmouse line, due to intermingled processes of proliferation, differentiation, andmigration in embryonic development. To study the migration of early neurons, inutero electroporation (IUE)wasused (seemethodsection fordetaileddescription).At E15, cells lining the ventricle were electroporated with different constructs, allexpressed under the control of the CAG promoter and with an internal ribosomalentry site (IRES) coupled GFP as a reporter protein. AnHCN‐DN subunitwith twomutations (G480A/G482A) introduced into theGYGmotif of thepore regionof thechannelwas used to ablate Ih (Mesirca et al., 2014). As control constructs, either awild‐type HCN subunit or a vector expressing only GFP was applied. With thisapproach, neurons from the VZ/SVZ at E15 were marked, which migrated intocorticallayersII/III.TheyreachedtheirdestinationatE19,thetimepointchosentoanalyzethismigrationassay.

Figure4‐8MigrationanalysisofIhablatedcellsinvivo

A,Representative images showE19murine cortices, thatwere inutero electroporated (IUE)atE15withGFP,wild‐typeHCN(HCN‐WT)withGFP,ordominant‐negativeHCN(HCN‐DN)alsowithGFPasreporter gene.All constructswere expressedunder the controlof theCAGpromoter.Most cells areable to migrate from the VZ/SVZ through the migration zone consisting of the intermediate zone,subplate,andlowercorticallayersintocorticallayersII/III.B,StereologicalquantificationoftheIUEanalysisshowsnosignificantdifferencebetweenthegroups.NumbersaregivenaspercentageoftotalGFP‐expressingcells.Scalebars,100µm;Dataispresentedasmean±S.E.M;One‐wayANOVA.n=4.

Results 4

49

By electroporation at E15, the cells in the VZ/SVZwere transfectedwith differentconstructs (GFP, HCN‐WT and HCN‐DN), and mostly migrated through theintermediatezone,subplate,andlowercortical layers,herereferredtoasmigrationzone, reaching cortical layers II/III at E19,without differences between the groups(Figure4‐8A).OnlyminorpopulationsweredetectedinthemigrationzoneandtheVZ/SVZ. Quantification did not show significant alterations between the groups(Figure4‐8B).

In summary, functionally Ih ‐ ablated cells showed no difference in their temporalpatternofmigrationintocorticallayersII/III.

4.4.4 ApoptosisinneonatalEMX‐HCN‐DNmice

Inadditiontotheimpairedproliferationanddifferentiation,anincreaseincelldeathdue to the ablated Ih and the altered biophysical properties of the transgene‐expressingcellscouldalsocontributetotheseveretheEMX‐HCN‐DNphenotype.Toinvestigatethispossibility,neonatalEMX‐HCN‐DNmicewereanalyzedforapoptoticprocesses. A staining for cleaved caspase 3, an enzyme whose activation inducesapoptosisuponcleavage,wasperformed.

4 Results

50

Apoptosis is a process needed for the refinement of neural connections and cellnumbersinthecerebralstructuresthatoccurphysiologicallyinthedevelopingbrain.In the regions analyzed, the sensorimotor cortex and the VZ/SVZ of control andmutant EMX‐HCN‐DNmice, only a small number of apoptotic cells, positive for thecleaved caspase 3 protein, were detected (Figure 4‐9 A). The stereologicalquantificationdidnotshowasignificantdifferencebetweenapoptoticcellnumbersofcontrolandmutantEMX‐HCN‐DNanimals(Figure4‐9B).

In thesensorimotorcortexandtheVZ/SVZof themurine forebrain,no impactof Ihablationonapoptosiswasdetected.

Figure4‐9ApoptosisinneonatalEMX‐HCN‐DNmice

A,Representative imagesofcleavedcaspase3staining(red) incoronalsectionsof thesensorimotorcortex and ventricular/subventricular zones (VZ/SVZ) counterstained with DAPI (blue) of control(con)andmutant(mut)EMX‐HCN‐DNmice. InmutantEMX‐HCN‐DNmice,GFPexpression isvisible.Insetsshowhighermagnificationoftheboxedareaswithsinglecellspositiveforcleavedcaspase3.B,Stereologicalquantificationdidnotrevealanysignificantdifferencesintheamountofapoptoticcellsintheregionsanalyzed.Datadepictedindensity(D)in[cell/mm3].Scale bars, 100µm;Data is presentedasmean± S.E.M; unpaired t‐test; cortex: con: n=7,mut: n=5;VZ/SVZ:n=6,mut:n=4.

Results 4

51

TheinvivophenotypeofIh‐deficientEMX‐HCN‐DNmiceexhibitingaseverereductioninthevolumeoftheforebrainstructuresduetoimpairedcorticaldevelopmentcanbeascribed to a reduction in proliferation and altered differentiation of the neuralprecursorcells.However,noeffectofIhdeficiencyonthemigrationorapoptosiswasdetected.

To confirm these results, and to dissect individual developmental processes fromeach other, cell culture experiments with primary cortical stem cells from the ratwereconducted.

4.5 InvitroexperimentsusingprimarycorticalstemcellswithpharmacologicalblockedIh

Cell culture experiments with primary cortical stem cells were designed incollaborationwiththeresearchgroupstemcellsof theDepartmentofNeurologyoftheUniversityHospitalandwereconductedbyDr.S.Vay.

The in vitro experiments allowed the different developmental processes to bediscriminatedfromeachotherandtoapplyadifferentapproachtotheablationofIhby the pharmacological blocker ZD7288. This chemical compound blocks HCNchannelswithoutdiscriminatingbetweenthedifferentsubunits,causingIhdeficiencyinthetreatedcells.

The proliferation and differentiation of primary cortical stem cells as the firstdevelopmentalprocessesinthecortexwereinvestigatedindifferentassays.

4 Results

52

Theproliferationofthecorticalstemcellswasinvestigatedbyanalyzingtheuridineanalog5‐bromo‐2’‐deoxyuridine(BrdU)incorporation,indicatingcellssituatedintheSphaseofthecellcycle(Figure4‐10A).Inaccordancewiththeinvivoexperiments,the invitromethodrevealedadecreaseinproliferationwhenthecellsweretreatedwith ZD7288 in a dose‐dependent manner (10 µM and 30 µM). This result wasreflected in theanalysesofdifferentiation,where theamountofSox2positivestemcellstreatedwith30µMZD7288wassignificantlydecreasedatday3,5and7afterFGF2 withdrawal, compared to control cells (Figure 4‐10 B). Withdrawal of FGF2resultedinthedifferentiationofthepluripotentstemcells.Atday10,theamountofSox2positivecellsalsosubstantiallydecreasedinthecontrolculture.Thus,therewasno longeranydifferencebetween thesegroups(Figure4‐10B). Incontrolcultures,the number of early neurons,markedwith TuJ1, increased between days 3 and 5,with a subsequent decrease in early neurons due to the culture conditions thatsupportedstemcellsbutnotneurons(Figure4‐10C).Atdays3and5theZD7288‐

Figure4‐10Invitro analysisofproliferationanddifferentiationinIh deficientcorticalstemcells

A,QuantificationofBrdUpositivecellsinrelationtototalcellnumberofcontrol(con)andZD7288(10µMand30µM)cellsrevealsadose‐dependentdecreaseinproliferation.B,C,D,E,Differentiationwasanalyzed3,5,7and10daysafterthewithdrawalofFGF2.B,Sox2,astemcellmarker,showsasignificantdecreaseofproliferatingcellsonday3,5and7ofZD7288 treatedcellscomparedwithcontrols.C,Earlyneurons,markedwithTuJ1aresignificantlyreducedincellnumbersonday3and5betweencontrolandZD7288treatedcells.D,UpontreatmentwithZD7288,astrocytes,stainedwithGFAP,appearedsignificantlyearlierandinhighercellnumberscomparedtocontrolcells.E,CNPasemarkedoligodendrocytesemergedearlieratday3whentreatedwithZD7288.*p<0.05,**p<0.01,***p<0.001,****p<0.0001;Dataispresentedasmean±S.E.M;A,one‐wayANOVAwithpost‐hocTukey’stest;n>200cellsperconditionfromonerepresentativeexperimentoutofthreeindependentexperimentswiththesameoutcome.B,C,D,E,Mann‐Whitneytest;B,C,n>500cellsperconditionanddayin2–6independentexperiments.D,E,n>300cellsperconditionanddayin2–5independentexperiments.(ExperimentswereperformedbyDr.S.Vay.)

Results 4

53

treatedcellsshowedsignificantlyreducednumbers inTuJ1positiveneurons,whichincreasedat days7 and10withno significantdifference left between control cellsand treated cells (Figure4‐10C).Glial cellsweremarkedwithGFAP for astrocytesand CNPase stained oligodendrocytes. In the control group, the amount of GFAPpositivecells increasedover time, reaching theirhighestexpression levelatday10(Figure 4‐10 D). In Ih blocked cells, GFAP positive astrocytes showed significantlyhighercellnumbersalmostthroughoutthewholetimepointsanalyzed(Figure4‐10D). The untreated control cells exhibited an increase in CNPase positiveoligodendrocytesuntilday7,followedbyadecreaseinCNPasepositivecellnumbersat day 10 (Figure 4‐10 E). The ZD7288 ‐ treated cells showed significantly highernumbersofoligodendrocytesatday3andremained increasedduringthe followingtime points except for day 7, but without significance compared to control cells(Figure4‐10E).

Both glial markers showed an early expression peak in ZD7288 ‐ treated cellscomparedtocontrolcells,whileTuJ1,amarkerforyoungneuronswasdecreased.

The invitro experiments resembled the resultsobtained from the invivoworkandshowedadecreaseinproliferationinIh‐ablatedcellsandanaltereddifferentiationpattern,asseenbytheearlyexpressionofglialmarkers.

Furthermore, migration and cell death in primary cortical stem cells withpharmacologicallyblockedIhwereinvestigated.

4 Results

54

To investigate the migration behavior of primary cortical stem cells, a TranswellmigrationassaywasperformedusingaBoydenchamber. In thisassay, cellshad tomigratethroughaporousmembraneinordertoreachthelowerpartofthechamber.Successfullymigratedcellswerelysedandtheopticaldensity(OD)ofthelysatewasmeasured.NodifferencewasdetectablebetweencontrolandZD7288‐treatedcells(10µmand30µm)(Figure4‐11A; left).Next, themigrationwasstimulatedbytheapplicationof10%FCStothetargetwell.Thisstimulusresultedinanoverallhighermigratory behavior, but did not show a significant difference between the groups(Figure4‐11A;right).

To exclude a toxic effect of the chemical compound ZD7288, which may causenecrosis or apoptosis, two different assays designed to analyze cell death wereperformed. Both methods were based on the detection of disruptions of themembrane, which mainly happens during necrosis, but also in late apoptosis(Darzynkiewicz et al., 1992;Chan et al., 2013). In the lactatedehydrogenase (LDH)assay,thereleaseofLDH,acytosolicenzymeuponmembraneleakage,wasmeasured.As a positive control, the supernatant of lysed cellswasmeasured, showing a highLDHlevel(Figure4‐11B).Treatmentwith10µMor30µMZD7288didnotresultina

Figure4‐11MigrationandcelldeathinIh‐ ablatedprimarycorticalstemcells

A,ATranswellmigrationassaywasperformedwithprimarystemcells.Nodifferencewasobservedbetweencontrol(con)andZD7288treatedcells(10µMand30µM(left)in theopticaldensity (OD)of the cell lysate, representing thecellsmigrated throughthe Boyden chamber. Upon stimulation of migration with 10 % FCS, the cellularbehavior did not alter between the groups.B, Lactate dehydrogenase (LDH) releasewasmeasuredasamarker forcelldeath, lysisofcells (positivecontrol)yieldedhighamounts of LDH but no difference between control and ZD7288‐treated cells wasdetected.C,Propidium iodid(PI) incorporation,amarker foradisruptedmembrane,wasmeasuredfordeadcellsandshowednodifferencebetweenthegroups.Dataispresentedasmean±S.E.M;A,n=2,onerepresentativeexperimentoutoftwoindependentoneswiththesameoutcome;B,onerepresentativeexperiment;n=4;C,n>1000cellsperconditioninoneexperiment.(ExperimentswereperformedbyDr.S.Vay.)

Results 4

55

difference in LDH release, compared to control cells (Figure 4‐11 B). Propidiumiodide(PI)isamembrane‐impermeabledye,thereforeintercalatingonlywithDNAofcellswithadisruptedmembrane.NodifferenceinthenumberofPI–negativeand,thereby, living cellswas detected between the groups (Figure 4‐11 C right). TheseresultsexcludeacytotoxiceffectofZD7288and,withit,Ihablationinprimarycorticalstemcells.

Despite thesmallnumberofexperimentalrepetitions, the invitroanalysisdoesnotsuggestaneffectofthemigratorybehaviorandcelldeathduetoZD7288application.Tovalidatetheseresultstheexperimentshavetoberepeatedmoreoften.

The in vitro experiments confirmed the results obtained by the in vivo methods,suggesting an impaired proliferation and differentiation of cells with attenuated Ihactivity. However, themigratory behavior and cell death were not affected by thepharmacologicalinhibitionofIh.

4.6 IdentificationofHCN‐dependentmechanisminproliferationanddifferentiationofneuralprogenitorcells

Thiswork is the first reportofHCN‐channelexpression in theembryonic forebrainandoftheimpactofHCN‐channelfunctioninneuraldevelopment.However,theexactmechanismsofhowthischannelanditscurrentcontributedtoearlydevelopmentalprocessesarenotyetknown.

To address the question of whether the observed phenotype was due to the non‐conductingchanneland,thereby,totheablatedIh,orwhetheritwasduetoanalteredinteraction of theHCN proteinwith other proteins, amouse linewith a shortenedintracellular HCN4 C‐terminus, resulting in the deletion of the cyclic nucleotide‐bindingdomain(CNBD),wasinvestigated(Aligetal.,2009).Fromtheliterature,itisknownthatmostproteinsinteractingwiththeHCNchannelbindtothelowerendoftheC‐terminus,wheretheCNBDregionis located(see1.3.1)(Wahl‐SchottandBiel,2009).Asthenameimplies,cAMPbindstotheCNBD,whichmeansthatthemutantchannel is insensitive to cAMP, potentially causing a hyperpolarized shift in theactivationcurve.

Inthemouselineanalyzed,ageneticapproachequivalenttotheEMX‐HCN‐DNmouselinewasused (seeFigure4‐2C). In theEMX‐HCN‐573Xmousemodel, the spatiallyrestrictedofexpressionwasachievedbytheEMX1promoter‐drivencrerecombinase,and additional temporal control was achieved using the Tet‐off system with theprepending floxed stop sequence. Expression of the HCN‐573X subunit and thereporterEGFPwasachievedusingthebi‐directionalpromoterPtet.

4 Results

56

Results 4

57

InEMX‐HCN‐573Xmice,amutationleadingtoaC‐terminaldeletionwasintroducedin the transgenicHCN subunit, resulting in a loss of theCNBD, themain region forinteractionswithotherintracellularproteinsandothermolecules(see1.3.3)(Figure4‐12A; for comparison see Figure 1‐2).Due to the absence of CNBD and impairedcAMP binding, the mutation is thought to result in a hyperpolarized shift of theactivationcurveatphysiologicalconditions.Ionfluxthroughthechannel,however,isnot affected (Schulze‐Bahr et al., 2003;Alig et al., 2009). In the adult forebrain, nogross morphological alterations were detected. The cortex developed normally,leavingonlythecolliculisuperioresandinferioresof themesencephalonuncovered(Figure4‐12B).Coronalsectionsof thebrainofEMX‐HCN‐573XmicewerestainedagainstNeuN(red),amarkerformatureneurons,revealingapartialoverlayofNeuNwith GFP positive transgenic cells in the dentate gyrus (Figure 4‐12 C; left panel;white arrows) and the striatum (Figure 4‐12 B; right panel; white arrows), whichdemonstrated theneuronal identity of amajor population ofHCN‐573X expressingcells. StainingwithGFAP, anastrocyticmarker (red), showedalso apartial overlaywith GFP‐expressing cells, mainly in the dentate gyrus (Figure 4‐12 D; left panel;whitearrows)andtheSVZ(Figure4‐1D;rightpanel;whitearrows), theregionsofneurogenesis in the adult brain. The identity of these GFAP/GFP positive cells isunclear. One possibility is that these cells are proliferative stem cells in the adultbrain located in the proliferative zones of the postnatal brain, the SVZ andsubgranular zone of the dentate gyrus (Doetsch et al., 1999; Doetsch, 2003). TheoligodendrocyticmarkerCNPaseshowednooverlapwithGFP, ineitherthedentategyrus(Figure4‐12E;leftpanel)orincortex(Figure4‐12E;rightpanel).Allsectionsanalyzed showed normal morphology with respect to the dentate gyrus of thehippocampus, the SVZ with adjacent striatum, corpus callosum, and lower corticallayers,aswellasuppercorticallayers.

Figure4‐12EMX‐HCN‐573Xshowednogrossmorphologicalalterationsintheforebrain

A,TheHCN‐572XmutationresultedinadeletionoftheC‐terminusattheendoftheC‐linker,causingthe loss of the CNBD (for comparison see Figure 1‐2).B,The brain of EMX‐HCN‐573Xmutantmiceshowednogrossmorphologicalalterations.C,Stainingofcoronalsectionswith theneuronalmarkerNeuN(red)showedanintactdentategyrus(leftpanel)andSVZwithpartsofthecorpuscallosumandstriatum.ApartialoverlapwithGFPpositivetransgeniccellsinthedentategyrusandthestriatumwasdetected(whitearrows).D,ForGFAPstaining(astrocytes;red), thesameregionswereanalyzedandresultedinapartialoverlaywithGFPpositivecellsinthedentategyrusandtheSVZ(whitearrows).D,CNPasestaining,amarker foroligodendrocytes(red), showed inadditionto the intactdentategyrus(left panel), no grossmorphological changes in theupper cortical layers (rightpanel).NooverlayofCNPaseandGFPpositive cellswasdetected inadultEMX‐HCN‐573Xmice. Small images showsinglechannelpicturesofGFPandtherespectivestaininginred.AllsectionswerecounterstainedwithDAPI(blue).Notethedifferentscalebars;Scalebars,100µm.

4 Results

58

Partial deletion of the C‐terminus, the most important domain concerning HCN ‐channel interactions with other molecules and proteins, did not cause amorphologicalphenotypecomparabletothatseeninEMX‐HCN‐DNmice.Thenormalmorphology found in these mice suggested that the interaction with most otherproteinsormoleculesisnotneededforthenormalproliferationofneuralstemcells.Furthermore, the impaired development of HCN‐DN ‐ expressing cells was notdependent on cAMP binding to HCN, and thereby independent of the resultingdepolarizedshiftofthechannelactivation.

Tovalidatethehypothesisthatimpairedproliferationoftheneuralprogenitorsisdueto the ablated Ih, electrophysiological voltage‐clamp recordings of the primarycorticalstemcellswereconducted.

Figure4‐13Voltage‐clamprecordingsofprimarycorticalstemcells

A,Sampletracesofavoltage‐clamprecordingshowanoutwardcurrentinresponsetonine500–ms‐ long depolarizing steps of 10 mV starting at ‐50 mV. The inset depicts the protocol used with apotentialof0mVmarkedinred.B,Sampletracesofavoltage‐clamprecordingshowafastinactivatingoutwardcurrentinresponsetoeleven500–ms‐longdepolarizingstepsof10mVstartingat‐80mV.Theinsetshowstheprotocolusedwiththe0mVpotentialmarkedinred.C,Sampletracesofavoltage‐clampexperiment recordedataholdingpotentialof ‐50mVwith2000–ms ‐ longhyperpolarizingsteps to a test potential of ‐120 mV shows no hyperpolarization ‐ activated current. The insetrepresentedthehyperpolarizingprotocolused.D,HCN4staining(red)ofstemcellsdepictedtheHCNchannelinintracellularcompartments.CounterstainwithDAPIisshowninblue.Scalebar,50µm.

Results 4

59

An electrophysiological investigation of the murine primary cortical stem cellsshoweddifferentoutwardcurrentsusingtwodepolarizingprotocols(Figure4‐13AandB). A slowly activating, non‐inactivating currentwas detected (Figure 4‐13A).Furthermore,afastactivatingandinactivatingoutwardcurrentwasmeasuredupondepolarization (Figure 4‐13 B). According to the ionic components andconcentrationsofthebathandelectrodesolutionsused(seematerialsection)andtheknownliterature(see1.2.1)(PickenBahreyandMoody,2002),theoutwardcurrentswerelikelymediatedbyapotassiumefflux.Thenon‐inactivatingcurrentpotentiallyresemblesadelayedrectifyingcurrent,while the inactivatingcurrentappeared likean A current. Confirmation of the exact channel population underlying thisconductancecanonlybeobtainedwithpharmacologicalexperimentsusingchannel‐specific blockers. Upon hyperpolarization, no current was detected and,consequentially, no Ih was detectable in these primary cortical stem cells. (Figure4‐13).Inlinewiththeseresults,stainingofthecellswithanHCN4antibodyrevealedthe internalization of the HCN protein, suggesting a non‐functional channelpopulation due to its internal localization (Figure 4‐13 D). HCN3 staining was notconductedinprimarycorticalstemcellsduetotheknownnonspecificstainingwiththeavailableantibodies.

The results obtainedwith the EMX‐HCN‐573Xmouse line (Figure 4‐13), suggestedthattheseveremorphologicalphenotypeseeninEMX‐HCN‐DNanimalswasnotdueto interacting proteins, or the depolarized shift of the activation curve upon cAMPbinding,butratherdependentontheablationofthecurrent.TheHCN4subtypewasdetected in primary cortical stem cells, but was found only in intracellularcompartmentsandnotinthecellmembrane(Figure4‐13D).Inagreementwiththisfinding, a current activated upon hyperpolarization was not detected in primarycortical stem cells (Figure 4‐13 C). Nevertheless, outward currents were detected,probablycarriedbypotassiumions,whichincludedadelayed,non‐inactivatingandafast,inactivatingcurrent(Figure4‐13AandB).

4.7 Summary

This work presents the first analysis of the impact of the HCN channel on earlyprenataldevelopmentoftherodentforebrain.Intheembryonicbrainofthemouse,HCN3 and HCN4 are the major subtypes found in different precursor cells andneuronsmediatingtheIh(Figure4‐1).Eveninhumancorticalstemcellsandneurons,theHCN3subtypewasdetectedatthemRNAlevel.Inlinewiththisearlyexpression,the ablation of the current by a genetic approach showed a severe reduction incortical and hippocampal structures in a development‐dependent manner (Figure4‐3,Figure4‐4).TheIhablationinvivoandthepharmacologicalblockinvitroshowed

4 Results

60

areductioninproliferation(Figure4‐5,Figure4‐6,Figure4‐10)andalteredtemporalpatternsofdifferentiation(Figure4‐7,Figure4‐10),withoutaffectingthemigrationorapoptosisofyoungneurons(Figure4‐8,Figure4‐9,Figure4‐11).Mechanistically,thephysiologicalproliferationwasnotinfluencedbythedeletionoftheC‐terminusofthe HCN subunit, which resulted in an impaired interaction with other interactingproteins or molecules (Figure 4‐12). The exact role of the HCN channel inproliferation could not be determined due to the absence of Ih in primary corticalstemcells,probablyduetointernalizationoftheprotein(Figure4‐13).However,thisis the first time that the impactof theHCNchannelsand themediatedcurrentwasdemonstratedinvitroandinvivoontheearlyembryonicdevelopmentofthemouseforebrain.

Discussion 5

61

5 Discussion

5.1 ExpressionpatternofHCNsubunits

The findings of this study suggest an important role of HCN channels in theproliferation of neural progenitor cells. Specifically the HCN3 mRNA shows apronouncedexpression in the telencephalonduringembryonicdevelopmentand indifferent neural progenitor cells (Figure 4‐1 A and B). Moreover, upon analysis ofmembranefractionsofE15brainlysates,theHCN3subtypewasalsodetectedontheprotein level. In addition toHCN3, a strongexpressionofHCN4mRNAandproteinwasdetectable(Figure4‐1AandD).AntibodystainingshowedthatHCN4islocalizedintheuppercorticallayersandco‐localizedwithTuJ1,aneuronalmarker,suggestingexpression inearlyneurons (Figure4‐1D)and, thus, anobvious spatialdistinctionfromHCN3 expression in neural progenitor cells. In contrast, theHCN1 andHCN2proteinswerenotdetectedintheembryonicmousebrain(Figure4‐1D).

As a result of the low abundance of HCN3 in the postnatal brain of rodents(Moosmangetal.,1999;NotomiandShigemoto,2004)anddifficultiesincloningandexpressioninheterologoussystems(Stieberetal.,2005;Mistríketal.,2006),littleisknown about the function and properties of this HCN channel subtype. This studypresentedhereindicatesanewfunctionoftheHCN3subtypeintheproliferationanddifferentiationofneuralstemandprogenitorcells.

To validate the observed expression pattern of the four HCN subtypes, publiclyavailabledatabaseswereanalyzedforHCNgeneexpression.

Consistent with the results in the present study, the Allen brain atlas of thedeveloping mouse brain showed in situ hybridization of the HCN1 mRNA andrevealed no expression at E11.5, E13.5, and E15.5 (www.developingmouse.brain‐map.org;Sep25,2016).Inthisdataset,theothersubtypeswerenotanalyzedatearlytimepoints.

Additionally, theGENSATorganization analyzed the expression ofHCN2 andHCN4subtypes(www.gensat.org;Sep25,2016).HCN1andHCN3expressionpatternswerenot investigated. The GENSAT organization generatedmice with the expression ofEGFPunder the controlofdifferentpromotersonabacterial artificial chromosome(BAC)integratedmultipletimesinthegenome.

According to the GENSAT data, the Hcn2 promoter is mainly active in thehypothalamus,midbrain,medulla, andpons.Within thecortical structures,HCN2 isonlysparselyexpressed in thesubplateandmarginalzone.ThisexpressionpatternsuggeststhelocalizationofHCN2inmigratinginterneurons.Thisinhibitorycelltype

5 Discussion

62

migrates tangentially between E15 and E16 from the medial ganglionic eminence(MGE) into theneocortex,guidedby themarginalzoneandthesubplate(Lavdasetal., 1999; Liodis et al., 2007). In the adult brain, several subpopulations ofinterneurons (e.g. somatostatin‐expressing neurons in the neocortex) expressfunctional Ih (Lupicaetal.,2001;Aponteetal.,2006;Maetal.,2006).Furthermore,theCajal‐Retziuscellsarelocalizedinthemarginalzoneoftheembryoniccortexand,thus, could potentially express HCN2. This transient cell population has a greatimpact on radial migration (Gil et al., 2014). Additionally, Cajal‐Retzius cells areknown to express Ih in the postnatal cortex (Kilb and Luhmann, 2000). The weakexpressionpatternshownbytheGENSATorganizationsupportsthepresentfindingsofHCN2expression.TheHCN2mRNAshowsincreasingamountsinthedevelopmentofthetelencephalon,withexpressionstartingintheembryo,whiletheHCN2protein,cannotbedetected in themembrane fractionofE15 embryos (Figure4‐1AandC).Furthermore, and again in linewith the GENSAT data showing theHcn2 promoteractivity in differentiated cells, the HCN2 subtype was not detected in the mRNAsequencingstudyrestrictedtoneuralprogenitorcells.

The Hcn4 promoter‐driven EGFP expression, investigated by the GENSATorganization,depictslocalizationinthemidbrain,medullaandponsaswellasinthedeveloping cortical structures atE15.5.Amoredetailed viewon the cortex revealsHcn4 promoter activity in the intermediate zone and cortical plate with higherexpressioninthesubplateandmarginalzone.ThisexpressionpatternsuggestsHCN4localization in pyramidal neurons migrating through the intermediate zone andcortical plate (see 1.1). Additionally, the higher expression in the subplate andintermediated zone can be explained by the localization of the thalamocorticalprojections,atthisearlytimepoint.ThalamocorticalneuronsareknowntoexpressIhintheadultmousebrain(Ludwigetal.,2003;Kanyshkovaetal.,2012).Thesubplateisneededtoformthesecorrectprojections(Ghoshetal.,1990;Hoerder‐SuabedissenandMolnár,2015).ThisexpressionpatterndiffersfromtheexpressionoftheHCN4subunitinthecorticalplatewhichwasfoundintheimmunohistochemicalanalysisofthisstudy.ApotentialexplanationforthisdiscrepancymightresideinthetransgenicBAC approach underlying the GENSAT data, which could be misleading as to theexpressionstrengthofindividualHCNsubunits.ThemultipleintegrationsoftheBACtransgene could lead to a much higher expression of EGFP compared to theendogenousexpressionoftherespectivepromoter(Chandleretal.,2007).Thiswouldexplain the broadEGFP staining in theGENSAT analysis,while the expressionwasrestricted to the marginal zone/layer I cells in the immunohistochemical analysis(Figure 4‐1 D). Additionally, the Ih was not detected in electrophysiologicalinvestigations of migrating neurons in the embryonic cortical structures (PickenBahreyandMoody,2002).Furthermore,theabsenceofHcn4promoteractivityintheventricular and subventricular zone (VZ/SVZ) is in agreement with the results

Discussion 5

63

presented here, suggesting the absence of HCN4mRNA in progenitor cells, mainlylocalizedintheVZ/SVZ.

Ontheprotein level,Seoetal.showedveryweakexpressionsof theHCN1,2and4subtypes in the intermediate zone of the emerging hippocampal anlage at E14.5,whilesuggestingthehighestexpressionforHCN4(Seoetal.,2015).Uponbroadeningof the analysis to the postnatal brain, the overall expression of the different HCNsubtypesinthehippocampusandcorticalplaterevealedadevelopmentalincreaseinHCN1 andHCN2 proteins and a decrease inHCN3 andHCN4; however, HCN3wasonlypartiallyanalyzed(Surgesetal.,2006;Kanyshkovaetal.,2009;Battefeldetal.,2012).

In accordance with the cortical expression patterns, analysis of different mouseembryonicstemcell linesshowedthepresenceoftheHCN3subtype,supporting itsroleintheproliferationofcells.Incontrast,muchlowerlevelsofHCN2mRNAwereobserved(Wangetal.,2005;Lauetal.,2011).

The exclusive expression of the HCN3 subtype in early neurons of the developingcortex (Figure 4‐1B) differs from previously experimental data and the literature,whichcollectivelysuggestedanexpressionoftheHCN4and,toalesserdegree,HCN2subtypesinearlyneuronsduringdevelopment.HCN4couldbedifficulttodetectduetotheguanine‐cytosine(GC)‐richsequences(Hcn4GCcontent:60%;NCBIReferenceSequence:NM_001081192.1). The difficulty to amplify theHCN4 transcript by PCRusingpolyA‐primer(personalcommunicationwithK.Sauter)suggestslongstretchesof GC‐rich sequences. To amplify the mRNA for the next generation sequencingperformedbyourcollaborators,thesamemethodwasapplied.ThiscouldresultinanunderrepresentationoftheHCN4mRNA.Moreover, thehigherexpressionofGFPintheGENSATdatasetduetopossiblemultipleintegrationsoftheBAC‐GFPsystemandthereby an artificial overactivity of the Hcn4 promoter could lead to anoverrepresentationoftheactualexpressionoftheendogenoussubunit.Moreover,thethalamocorticalprojections,possiblyalsoexpressingHCN4,canmimictheexpressioninthesubplate/intermediatezoneduetothecloselocalizationandthelowresolutionof the analyzed imagesby theGENSATorganization.Altogether this can result in anon‐physiological representation of the HCN4 subtype expression pattern, whichneedsfurtherinvestigation.

Analysisof thehumancell types investigatedbymRNAnext‐generationsequencingshowedagradualincreaseoftheHCN3expressionwhencomparingaRGstobRGstoneurons(Figure4‐1Blowerpanel),suggestingadifferentialimpactofthechannelonthese distinct cell types. In the course of the evolution, the impact of the basalprogenitors on the brain has been increasing, contributing to amore complex andbiggercortex(Molnár,2011);whilenobasalprogenitorsexistinthereptilebrain,the

5 Discussion

64

above‐described development of themurine brain showed evolving populations ofthebasalcelltypes,whiletheprimatecerebralcortexisevenmoresocharacterizedby a proliferative and large basal progenitor pool (Molnár, 2011). This in turnsuggestsanimportantroleoftheHCNchannelsduringbraingrowthinhumans.

Insummary, thefindingspresentedheresuggestadevelopmentalswitch incorticalexpression of the specific HCN subtypes, with HCN3 and HCN4 being mainlyexpressedduringembryonicphases,whileHCN1andHCN2beingexpressedlaterinthe postnatal and adult stages. Furthermore, theHCN3 subtype is expressed in theproliferativepopulationofneuralstemandprogenitorcells,whileHCN4isexpressedinyoungneurons.

5.2 EMX‐HCN‐DNmouselineasmodelforHCNchanneldysfunctionduringembryonicbraindevelopment

To investigate theprenatalroleofHCNchannelsand Ih invivo, theEMX1promoterwasselectedforspatialandtemporalrestrictionoftheHCN‐DNexpression.EMX1isexpressedfromE9.5on, ina forebrain‐restrictedmanner inexcitatoryneuronsandpartially in glia cells, corresponding to the cell types generated in theVZ/SVZ (See1.1; Figure 4‐2) (Iwasato et al., 2000; Gorski et al., 2002). To verify the reportedforebrain restriction, EMX1‐Cre mice were bred the with the Ai9 reporter lineexpressingtheredfluorescenttandemdimer(td)Tomato(Figure4‐2B).TdTomatoisalargecytosolicprotein.Thismouselineshowedtheexpectedforebrainrestrictionandthereporterwasnotexpressedinregionslikethecerebellum(Figure4‐2‐B).Inthedentategyrus,tdTomatowasnotdetectedinthesubgranularzoneorthesomaofCA3 pyramidal neurons, while the CA3 pyramidal cell axons showed a redfluorescence (Figure 4‐2‐B). This distribution of tdTomatowas also seen inmouselines using different promoters with similar expression patterns (like CaMKIIα)shownintheAllenbrainatlas(mouseconnectivitystudies;www.connectivity.brain‐map.org/transgenic; Sep 25, 2016), suggesting the localizationmainly in dendritesandaxons.

The dominant‐negative HCN subunit, expressed as a transgene in the mouse linesfurtheranalyzedinthisstudy,interactswithendogenousHCNsubunitsandresultsina functional ablation of Ih in a subunit‐unspecific manner. The dominant‐negativeeffectofHCNsubunitswithcurrent‐inhibitingmutations,mainlyintheGYGmotifofthe pore region, was shown before (Xue et al., 2002; Er et al., 2003), while thefunctionalityof theHCN‐DNmutationused in thisworkwas initiallyestablished inourlaboratory(Sandke,2006;Merseburg,2016).Thesubunit‐unspecificablationofIhcanpreventthecompensatorymechanismleadingtoup‐regulationofunaffectedHCN

Discussion 5

65

subunits in single knockout studies. Analyzing mouse lines with individual HCNsubtype knockouts showed morphologically normal brains, suggesting thesecompensatorymechanisms(see1.3.4).

The genetic approach used in the EMX‐HCN‐DN mouse line additionally tookadvantageoftheTet‐offsystem.Uponcrerecombinaseactivationofthetetracyclinetransactivator(tTA)andintheabsenceofdoxycyclinethetTAbindstoitsresponsiveelement(Ptet)enablingthetransgeneexpression(Figure4‐2C).Theadministrationofdoxycycline thus leads to the inactivation of the transgene expression, which is acontrol to exclude positional effects by the manipulated genome of the transgenicmice (Eszterhas et al., 2002).While the transgenic animals that arenot exposed todoxycycline exhibit severe reductions of body size and brain volume, doxycycline‐treated animals show a normal body weight and size as well as normal forebrainmorphology (Figure 4‐3). These findings exclude a phenotype‐causing positionaleffectoftherandomlyinsertedtransgenicHCN‐DNconstruct.

InthetransgenicEMX‐HCN‐DNmice,thereductionofbodysizeandweighttogetherwith the reduction in forebrainstructures (Figure4‐3) led toan increased lethalityaroundweaning precluding analyses in adult animals. Therefore, juvenile (3‐week‐old) mice were characterized instead. It was beyond the scope of this study toinvestigatethemechanismsunderlyingthisphenotype.

Neonatal and juvenile EMX‐HCN‐DN mice showed a significantly decreased brainvolume compared to control animals and compared to mutant mice exposed todoxycycline (Figure4‐4C).Additionally,mutantmiceatP21showeda significantlyincreased ventricle volume compared to the control and doxycycline‐treated group(Figure 4‐4 F). This severe volume increase is due to the small remaining cortexincompletely covering diencephalon and mesencephalon, and leaving the lateralventriclesopentothesubarachnoidspace.

Next, the mechanisms underlying the severely reduced forebrain structure wereexamined, focusingonkeydevelopmentalprocesses suchasneuronalproliferation,differentiation,andmigration.Whilethedifferentiationispostnatallydetectabledueto the late expression of glialmarkers (see control animals in Figure 4‐7) and themigration was analyzed at embryonic ages (Figure 4‐8), the proliferation, mainlytaking part at embryonic time periods (E11 ‐E16, (Kwan et al., 2012)), was alsoanalyzed postnatally (Figure 4‐5). The proliferation zones VZ/SVZ were clearlydistinguishableby immunohistochemistrywith aKi67antibody in themutantmice(Figure4‐5).Withthisclearidentificationoftheremainingareaofproliferationintheneonatal brain, this process could be analyzed. For a more detailed prenatalinvestigation of the proliferation of Ih ablated cells, earlier embryonic time pointsshouldbeanalyzedinthefuture.

5 Discussion

66

In summary, themorphological alterations seen in theEMX‐HCN‐DNmice are verylikelyattributable to thedominant‐negativeHCNsubunit, i.e.due to the Ih ablation.Themouse linegave theopportunity to targetand investigate theproliferationanddifferentiation of Ih ablated neural stem and progenitor cells during corticaldevelopmentinvivo.

5.3 TheeffectofIhblockadeonproliferation

The key findings of this work revealed the impact of genetic or pharmacologicalinhibitionof Ih on theproliferationofneural stemandprogenitorcells. InneonatalEMX‐HCN‐DN animals, the proliferation in the VZ/SVZwas significantly reduced invivo (Figure4‐5). Inagreementwiththisresult, the invitroexperiments inprimaryrat cortical stem cells showed a decrease in BrdU incorporation upon ZD7288treatment (Figure 4‐10). Furthermore, the link between HCN‐DN expression,impairedproliferationandthereducedforebrainsizeofthetransgenicEMX‐HCN‐DNanimalswas corroborated by the absence of structural changes in the forebrain ofNEXCre‐HCN‐DNanimals (Figure4‐6), inwhich the onset ofHCN‐DNexpression isdelayedtopost‐mitoticneurons.

InhibitionofIhleadstoahyperpolarizationofthemembranepotentialintheHCN‐DNexpressing transgenic cells aswell as in ZD7288 treated cells (Aponte et al., 2006;Merseburg,2016).Hyperpolarizationofthemembranepotentialofneuralstemcellsmay interfere with their proliferation. Two distinct mechanisms could account forthiseffect:ontheonehandtheinfluenceoftheaveragerestingmembranepotentialofneuralstemcells,andontheotherhandmembranepotential fluctuationsduringcellcycleprogressioncouldbedependentontheIh.

5.3.1 RestingmembranepotentialinneuralstemcellsandthecontributionofIh

The membrane potential, determined by the concentration gradient of extra‐ andintracellular ionsandsetby theactivityofdiverse ionchannelsandpumps, is longknown to influence crucial cellular processes. The average resting membranepotential possibly has an influence on the proliferation of cells. The membranepotentialprogressivelyhyperpolarizesduringcelldifferentiation.Whilethefertilizedegg and embryonic stem cells are in a depolarized state, differentiated, non‐proliferative cells are hyperpolarized (Binggeli and Weinstein, 1986; Yang andBrackenbury, 2013). However, depending on the proliferative state of the cell, onespecificcelltypecanexhibitdifferentmembranepotentials.Forexample,theChinesehamsterovary(CHO)cellshaveamembranepotentialofapproximately‐60mVwhentheyareinaquiescent,confluentstate(ConeandTongier,1973).Upondissociation

Discussion 5

67

themembranepotentialdepolarizesbyabout50mVtoapotentialofapproximately‐10 mV, resulting in the proliferation of CHO cells (Cone and Tongier, 1973).Furthermore,thedepolarizationofthemembraneofproliferativecellswasshowninpulmonarysmoothmusclecells(Platoshynetal.,2000).Ahyperpolarizingshiftofthemembrane potential in human mesenchymal stem cells during adipogenic orosteogenic differentiation was detected (Sundelacruz et al., 2008). Upon artificialdepolarization by increased extracellular potassium concentration or ouabain, thedifferentiationarrestedasseeninthesuppressionoftheexpressionofgenesmarkingbothdifferentiationprocesses(Sundelacruzetal.,2008).Interestingly,theattempttoreverse the differentiation of osteoblasts and adipocytes, upon depolarizationresultedinadownregulationoftheexpressionofdifferentiationmarkergenes,whilethe upregulation of stem cellmarkerswas not detected (Sundelacruz et al., 2013).Furthermore, neurons show the most hyperpolarized membrane potential in themammalian system hence lack the ability to proliferate. Interestingly, culturedneurons regain their mitotic activity upon long lasting pharmacologicaldepolarization,yetlackingthefinalcytokinesis(ConeandCone,1976).Inthisstudytheharshtreatmentofthematureneuronswithouabain,asodium/potassiumpumpinhibitor(Laursenetal.,2013),veratridineasnon‐specificsodiumchannelactivator(Yoshinaka‐Niitsu et al., 2012), or the cation‐selective pore forming proteingramicidine(Hicksetal.,2008)canhaveavarietyofsecondaryeffects,onebeingthemisguidedmitoticabilityoftheneurons.Nevertheless,consistentwiththisresult,theoptogenetic depolarization of youngmigrating neurons in acute embryonic corticalslicesledtothere‐entryinthecellcycleandtocelldivision(CalderondeAndaetal.,2016). In contrast to thedifferentembryonic stemandprogenitorcells, the restingmembrane potential of adultmurine neural progenitor cellswas hyperpolarized atapproximately –80mV, but the enhanced proliferation seen upon barium chlorideapplicationwasassociatedwith thedepolarizationof themembrane (Yasudaet al.,2008).Thehyperpolarizationintheproliferativestatewasalsoobservedinpostnatalastrocyticprogenitors(MacFarlaneandSontheimer,2000).

As mentioned above, the HCN channels contribute via hyperpolarization‐activateddepolarization, to the establishment of the membrane potential and are therebycrucial for the determination of the biophysical properties of the cell. In matureneurons,theexpressionoftheHCN‐DNsubunitleadstoahyperpolarizedmembranepotential compared to control cells (Merseburg, 2016) due to the loss of sodiuminflux through HCN channels close to the neuronal resting potential. The reversalpotential of the ion flux through different HCN channels is around ‐25 to ‐40 mV(Mistriketal.,2005;RodriguesandOertel,2005;Stieberetal.,2005),which is stillmoredepolarizedthanthemembranepotentialofneuralstemcells,beingaround‐40to ‐70 mV (Aprea and Calegari, 2012). The Ih is thereby potentially driving thedepolarizationofthemembrane.

5 Discussion

68

BasedontheknownpropertiesofIh,thechannelpresumablydoesnotexhibitahighopenprobabilityatthemembranepotentialofrestingneuralstemcells.Accordingtothe literature,HCN3andHCN4channelsstart toactivateataround‐70mVand‐80mV,respectively,andthevoltagesofhalfmaximalactivationsoftheHCN3andHCN4subtypesinHEKcellsare‐95+/‐1mVand–109+/‐1mV,respectively(Ludwigetal.,1999; Mistrik et al., 2005). However, the human HCN3 and HCN4 channels showalreadyasmallchannelopenprobabilitystartingatamembranepotentialaround‐50mV(Stieberetal.,2005;Bieletal.,2009),whichmayresultinasmallcurrentattherestingmembranepotentialofneuralstemcellsand,thus,mightactdepolarizingonthesecells.Thesepropertiesarehighlydependentontheexperimentalsetupandcanvary in the physiological setting of the stem cells. Furthermore, Ih can stabilize themembranepotentialatthisdepolarizedstateduetothepreviouslyreportedvoltage‐independentcurrentIins(Proenzaetal.,2002;Aponteetal.,2006).Moreover,theco‐expressionandinteractionofaccessorysubunitsormodulatingmoleculesareabletoshifttheactivationthresholdtomoredepolarizedpotentialsandcouldcontributetoahighercurrentdensityattheneuralstemcells’restingmembranepotential(Zollesetal.,2006;Lietal.,2009;Santoroetal.,2011).

TheresultspresentedherearesupportedbyexperimentsregardingtheimpactofIhinhibitionontheproliferationinamouseembryonicstemcellline(Lauetal.,2011).Theelectrophysiological investigationshowedahyperpolarization‐activatedcurrentbut did not unambiguously identify this current as Ih given that the application ofZD7288 resulted only in a partial inhibition (Lau et al., 2011). Nevertheless, inaccordancewithWang et al. the expression of the HCN2 and HCN3 subunitswereverifiedbyPCR,whileonlytheHCN3proteinwasdetectedintheplasmamembraneofthestemcells,explainingthecAMPinsensitivityoftheIhmeasuredbyWangetal.(Wangetal.,2005;Lauetal.,2011).Uponcurrent inhibitionbycesiumchlorideorZD7288 theproliferationwasdecreased in a dose‐dependentmanner, showing thesameeffectasthatdetectedininvivoandinvitroexperimentspresentedhere.

Together, the data obtained by genetic or pharmacological inhibition of Ih stronglysupportthehypothesisofhyperpolarizationofthemembranefollowingattenuationof HCN channel function, which resulted in the inability of progenitor cells toproliferate(Figure4‐5andFigure4‐10).

5.3.2 MembranepotentialfluctuationsduringcellcycleprogressionandthepotentialcontributionoftheIh

Inadditionto theoveralldepolarizedmembranepotential, the literaturesuggestsaslowoscillationofthemembranepotentialduringcellcycleprogression(Blackistonet al., 2009; Urrego et al., 2014). The S‐phase of the cell cycle is associatedwith arelativehyperpolarizationofthemembranepotential,whilethemembraneofcellsin

Discussion 5

69

M and G1 phases is more depolarized (Sachs et al., 1974; Boonstra et al., 1981;Wonderlinetal.,1995;Ngetal.,2010;Lauetal.,2011;Urregoetal.,2014).Theexacttimecourseoftheriseandfallofthepotentialseemstodifferbetweencelltypesandexperimental settings (Sachs et al., 1974; Boonstra et al., 1981). Additionally, thefluctuationappearstodifferwithinpostnatalastrocytes,resultinginadepolarizationupon S‐phase entrance (MacFarlane and Sontheimer, 2000). Taken together thesefluctuations arewell established in cancer cells andotherhighlyproliferative cells.Althoughthese findingshavenotbeen investigatedinneuralstemcellsso far, theiruniversal validity is widely assumed and there are studies suggesting a similarmechanism of cell cycle regulation in neural progenitor cells. In embryonic ratoligodendrocyteprogenitors,themembranepotentialoftheG0/G1phasewasmoredepolarized compared to the hyperpolarized S phase (Ghiani et al., 1999).Interestingly the inhibition of potassium channels or the activation of sodiumchannels,bothcausingdepolarization,inhibitthecellcycleintheG1phase(Ghianietal., 1999). The application of both substances only led to the inhibition ofproliferation when applied during the G1 phase, showing the necessity of thehyperpolarizationforthecellcycleprogressionintotheSphase(Ghianietal.,1999).Furthermore, thedepolarizationduring theMphasewasshown ingliomacellsandproliferating astrocytes (Habela et al., 2008). Collectively, these results suggest thepresenceofthemembranepotentialoscillationsalsoinneuralprogenitorcells.

Inmurineembryonicstemcells,thesecellcyclefluctuationswerealsodetected(Ngetal., 2010; Lau et al., 2011). Interestingly, closer investigation of the cell cycle uponHCNchannelinhibitionwithZD7288revealedadecreaseinthenumberofcellsintheG1phase,whilethecellnumberintheSphaseofthecellcyclewasincreased(Lauetal.,2011).Inanothermurineembryonicstemcellline,theseresultswerereproduced,and a further analysis of the cell cycle in ZD7288‐treated cells was conducted(Omelyanenkoetal.,2016).Theentirecellcyclewasprolongedbyapproximately30%.ContrarytoLauetal.,theyhaveshown,thatthenumberofcellsintheSphasewasincreasedZD7288treatment,butthecellnumberinG2/Mphasewasdecreased,withno difference in the G1 phase (Lau et al., 2011; Omelyanenko et al., 2016).Omelyanenkoetal.calculatedthelengthofthedifferentcellcyclephasesconsideringthe doubling time and cell cycle fraction. Their calculation indicated that ZD7288treatmenthadincreasedG1andSphasesduration,whileonlytheSphaselengthenedinthestudyconductedbyLauetal.(Omelyanenkoetal.,2016).Thisdifferenceinthelengthofcellcyclephasescouldbeduetothedifferentcell linesand/orintroducedbythedifferentcultureconditions(Jovicetal.,2013;Tammetal.,2013).

Furthermore,detailedanalysisoftheSphaserevealedaslowerprogressionthroughthis cell cycle phase upon Ih inhibition, by the slower incorporation of modifiednucleotidesintotheDNA(Omelyanenkoetal.,2016).

5 Discussion

70

Preliminarydata from this study indicatean increase in the lengthof the cell cycleand S phase in the rat primary neural stem cells of treated with ZD7288 byinvestigationofBrdUincorporationovertime(datanotshown)(Nowakowskietal.,1989).

According to these results, one might speculte that the Ih is activated upon thehyperpolarization at the S phase and slowly depolarizes the membrane potential,whichisneededfortheentryintotheG2phase.UponIhinhibitiontheSphaselengthincreases, until the cell cycle progression stops at the end of the S phase andproliferationisblocked.

In summary, the reduced proliferation in the Ih‐ablated cells in vivo and in vitro islikelycausedbythealteredbiophysicalpropertiesoftheneuralstemandprogenitorcells,leadingtoahyperpolarizationofthemembranepotential.Inneuralstemcells,therestingmembranepotentialisbetween‐40mVand‐70mV(ApreaandCalegari,2012). The cells exit the cell cycle by shifting tomorehyperpolarizedpotentials. Amoredetailedinvestigationofthemembranepotentialshowsoscillationsduringcellcycleprogression,whichareslowandassumedtobestillinamoredepolarizedstatethan the resting membrane potential of the corresponding non‐proliferative cells(YangandBrackenbury,2013).ThehyperpolarizationisneededtoentertheSphasewhile a slow and progressive depolarization is suggested until the depolarizedmembranepotentialintheMphaseisreached.UponIhablation,itislikelythatcellsprogressmore slowly through the S phase and are not able to enter theG2phase,resultinginacellcycleexitandadecreaseinproliferation.

5.4 Impactofthemembranepotentialonproliferation

The abovementioned assumptions suggest a role of the Ih in the proliferation ofneural stem and progenitor cells mainly due to its influence on their biophysicalproperties,i.e.bymodulatingtheirmembranepotential.Ithaslongbeenknown,thatthemembranepotentialiscloselyconnectedtotheproliferationofdifferentcelltypes(Cone and Cone, 1976). Even though, the direct connection is not clear yet. Thefollowingchapter,therefore,isabouttospeculateontheinfluenceofthemembranepotentialonproliferation,itsconnectiontoneuralstemcellsandthepossibleimpactoftheIhontheseprocesses.

5.4.1 Calciumsignaling

Calcium is one of the most important second messengers in cells and contributesfundamentally to the cell cycle progression and thereby to the proliferation of avariety of cells (see 1.2.1) (Capiod, 2011). The expression of different types of

Discussion 5

71

voltage‐gated calcium channels (VGCC) was shown in embryonic stem cells anddifferent cancer cells (Rodriguez‐Gomez et al., 2012; Gray et al., 2013). On thecontrary,theinfluxofcalciumthroughVGCCsisofminorimportanceinneuralstemcells. Most studies suggested a role of the VGCCs in the differentiation of neuralprogenitor cells (D’Ascenzo et al., 2006; Lepski et al., 2013; Toth et al., 2016).However, there are reports about calcium currents in the embryonic neuralprogenitorcells,althoughtheydonotaccountforthepossiblepresenceofcellsthatundergo differentiation (LoTurco et al., 1995; Cai et al., 2003; Picken Bahrey andMoody,2003).

This suggests that the cytosolic calcium increase observed in proliferating neuralstem cells is mediated by internal calcium stores or ligand‐gated channels andtherebyindependentofthedepolarizingactionoftheHCNchannelsandthemediatedcurrent(see1.2.1).

5.4.2 Directactivationofcellsignalinguponchangesinmembranepotential

Theactivationofvariousproteinscanbedirectlylinkedtothedepolarizationoftheplasmamembrane.

The nano‐clustering of K‐Ras was shown upon membrane depolarization byincreasing extracellular potassium concentration (Zhou et al., 2015). The nano‐clustering leads to the activation of the extracellular signal‐regulatedkinase/mitogen‐activated protein kinase (ERK/MAPK) pathway (Zhou et al., 2015).TheMAPkinaseactivationcanbedirectlylinkedtotheproliferationofdifferentcelltypes(Dhillonetal.,2007).

Furthermore,thenuclearrespiratoryfactor(Nrf)2isatranscriptionfactorregulatinggeneexpressionassociatedwithDNAreplication,mitosis,andcytokinesis(Scarpulla,2008). Interestingly, its activation by the translocation to the nucleus can beprovokedbydepolarizationoftheplasmamembrane(Yangetal.,2004),leavingNrf2asaperfectcandidatetodirectlyconnectthemembranepotentialandproliferation.

Additionally, there are a number of immediate early genes that are induced uponmembranedepolarizationandthoughttobeinvolvedinneuralplasticity(Herschmanet al., 2000).Thekinase inducedupondepolarization (KID)1 is one such example.KID1 is found in the brain and conducts auto‐phosphorylation and histonephosphorylation,therebyregulatinggeneexpression(Feldmanetal.,1998).

The potassium channel ether‐à‐go‐go (EAG), which can be found in a variety ofdifferentcancertypes,promotesproliferation(Pardoetal.,1999;PardoandStuhmer,2008). Interestingly, the current itself only partially leads to the increase in

5 Discussion

72

proliferation. Additionally, the conformational change upon voltage sensing, evenwithoutcurrentflow,elevatesproliferationactivityintheaffectedcells(Hegleetal.,2006). Hegle et al. used amutation in the selectivity filter of the EAG1 channel toshowthecurrent‐independentfunctionofthechannel(Hegleetal.,2006).TheEMX‐HCN‐DNmicealsohaveamutationintheselectivityfilter.Thismutationblocksthecurrentbutdoesnotchangetheconformationuponvoltagesensing,arguingagainstthecurrent‐independentfunctionseenintheEAGchannel.

Insummary,thedepolarizingactionand/ormembranepotentialstabilizationoftheHCNchannelsandthemediatedIhcanfunctionasanindirectregulatorofproteinortranscriptionactivation.

5.4.3 Cellvolumeregulation

Thealterationsofthecellvolumecaninfluenceimportantprocesseslikeproliferationor apoptosis (Lang et al., 2007).Becauseof the tight couplingof the volume to theintracellular ionconcentrations, it isdifficult todistinguish theirrespective impactson membrane potential regulation and cell volume. It is known that the volumealternates during cell cycle progression, being increased at the beginning of theMphase but shrinking toward its end, accompanied by chloride and potassium flux(Langet al., 2007;Hoffmannet al., 2009).The smallestvolume ismeasuredduringtheG1/Sphase,whenintracellularcalciumandsodiumconcentrationsarepotentiallyhigher, and in turn can lead to the subsequent volume increase (Lang et al., 2007;Hoffmannetal.,2009).

DuetothesodiumconductanceoftheHCNchannels,thesechannelscouldcontributetothesodiuminflux,andtherebytothevolumeincrease.

TakentogethertheHCNchannelsandtheirmediatedIhcanhavedualfunctionsintheproliferation activity of neural stem and progenitor cells, by influencing themembranepotentialandbyitssodiuminwardcurrent.

5.5 TheinhibitionofHCNchannelsanditsinfluenceondifferentiation

The data obtained in this study revealed that the genetic or pharmacologicalinhibitionofHCNchannelsalteredthedifferentiationofneuralstemandprogenitorcells.Both invivoand invitro findingsshowedanearlyappearanceofglialmarkerssuggestingatemporalchangeinthedifferentiation(Figure4‐7andFigure4‐10B,C,D,E).Apreliminaryexperimentinvestigatingtheratioofprimarycorticalstemcells

Discussion 5

73

which incorporated BrdU indicated that the primary neural stem cells have anincreaseincellcyclelength(datanotshown)(Nowakowskietal.,1989).

Inagreementwiththiswork,inhibitionofIhwithZD7288inmurineembryonicstemcells led toadecrease inpluripotencyandan increaseofdifferentiationmarkers inspontaneousdifferentiatingconditions(Omelyanenkoetal.,2016).Togetherwiththedecreaseinproliferation,thiswasaccompaniedbyalengtheningofthecellcycle(Lauet al., 2011; Omelyanenko et al., 2016). Omelyanenko et al. showed that thelengtheningwas due to an increase in G1 and S phase length (Omelyanenko et al.,2016).Fromgeneralstemcellresearch,itisknown,thatthelengthoftheG1phaseisanimportantparameterforthedifferentiationofacell(Chenetal.,2015).DuringtheG1 phase, the cell is susceptible to differentiating factors. Thus, G1 phase inpluripotentcellsisfast,andalongerduringdifferentiationdivisionsresultinginafastphaseinpluripotentcellsandlongerphaseduringdifferentiationaldivisions(Chenetal.,2015).Thiswasalsoshowninneuralstemcells(Calegarietal.,2005;Lukaszewiczet al., 2005; Roccio et al., 2013). Furthermore, experiments in neural stem cellsrevealedthatthisisnotonlyconsequence,butalsocauseforthedifferentiation.Theartificial lengthening of the G1 phase induced and increased neurogenesis andgliogenesis,while proliferative divisionswere decreased (Lukaszewicz et al., 2002;Calegari andHuttner, 2003; Roccio et al., 2013). Furthermore, the cell cycle lengthincreased during cortical development as a function of time, independent of theincreasedneurogenicdivisions(Takahashietal.,1995;Calegarietal.,2005).

Thelengtheningofthecellcycle,andinparticularthelongerG1phaseintheZD7288‐treated cells, would lead to an earlier differentiation of the neural stem cells. Inagreementwiththisassumption,theinvitrodifferentiationassayshowedadecreasein Sox2 positive stem cells upon Ih inhibition (Figure 4‐10 B). However, themajorfindingof the invivo and invitro differentiationexperimentswas that theZD7288‐treatedcellculturesexhibitedtheglialcellfatemuchearlier,asevidentbytheearlyexpressionofglialmarkers.Thisfindingwouldbeexplainedbythelengtheningofthecellcycleduringthecorticaldevelopment leadingtoamuch longercellcycle in thelater gliogenesis compared to the earlier neurogenesis (Takahashi et al., 1995;Calegari etal., 2005). In summary, the lengtheningof the cell cyclewould lead toadecrease in proliferative divisions and an increase in differentiation potentiallytowardstheglialcellfate.

Afurtherexplanationforthetemporalalterationsinthedifferentiationoftheneuralstem cells in vivo and in vitro can be related to voltage‐gated calcium channels. Inmurineneuralstemandprogenitorcells,smallL‐typecalciumcurrentsweredetectedwithadifferentiation–dependentincreaseincurrentdensity(D’Ascenzoetal.,2006).InhibitionoftheL‐typecalciumchannels ledtoadecreaseof theneuronaloutcome(D’Ascenzoetal.,2006;Lepskietal.,2013).Onthecontrary,itsactivationincreased

5 Discussion

74

the cell numbers with neuronal fate. Both treatments had no effect on theproliferationoramountofanalyzedcells(D’Ascenzoetal.,2006).IncreasedlevelsofcAMPandthesubsequentactivationoftheL‐typecalciumcurrentshowedthesameeffectsontheneuronaloutcomeoftheneuralprogenitorcells(Lepskietal.,2013).Inboth studies, it was not investigated if the decreased neuronal cell fate due toinhibitionoftheL‐typecalciumcurrentwasaccompaniedbyanincreaseinglialcells.

Furthermore,inHEKcellsandhippocampalneuronsLinetal.wereabletoshowaninhibitionof theL‐type calciumchannelsby the interactionwith theN‐terminusofthe closed HCN2 subunit (Lin et al., 2010). This suggests that the altereddifferentiation in theEMX‐HCN‐DNmicewouldbedue to theoverexpressionof theHCNproteinandtherebyanincreasedinteractionbetweentheHCNN‐terminusandtheL‐type channels. Furthermore, the altereddifferentiationwouldbe seen also inthe EMX‐HCN‐573Xmice,with the deleted C‐terminus, and in the IUE experimentsuponelectroporationand therebyoverexpressionof theporemutantand thewild‐typeHCNchannel,butwerenotfurtherinvestigated.However,itdoesnotexplainthealtered differentiation outcome of the neural stem cell culture treated with Ihinhibitor andexpressingonly endogenous levelsofHCNprotein.Togetherwith thediverging sequences of the N‐terminus of the different HCN channel subtypes thisinteractionisnotverylikelybutcannotbeexcluded.

Takentogether,thealtereddifferentiationpatternoftheIh‐blockedcellsinvivoandinvitroislikelyduetothecellcyclelengthening,resultinginanearlydifferentiationoftheneuralstemandprogenitorcells.Furthermore,thehyperpolarizingshiftoftherestingmembranepotentialuponIhinhibitionshiftsthemembranepotentialfurtherawayfromtheL‐typecalciumchannelactivation.Theshiftwould,inturn,reducethecalciuminfluxtroughL‐typeVGCCintothedifferentiatingcell,inhibitingtheneuronalcellfateandpotentiallyincreasingglialcellnumbers.

5.6 HCNchannelsinteractionswithotherproteinsandsmallmolecules

HCN subunits can interactwith a variety of different proteins and smallmoleculeswhichmodulateIh(see1.3.3).ByinvestigatingtheEMX‐HCN‐573Xmouseline,theseinteractionswerepartiallyanalyzed.TheexpressionoftheHCN‐573Xsubunit,whichlackedtheCNBD,resultedinanormalmorphologyoftheanalyzedforebrains(Figure4‐12). The overexpression led to the heteromerization of the endogenous subunitswith the mutated HCN‐573X, resulting in a dominant‐negative effect on themodulationofthechannel(Aligetal.,2009).Furthermore,theCNBDisneededfortheinteractionoftheHCNchannelswithcAMP,TRIP8b,MiRP1,KCTD3,filaminA,tamlin,

Discussion 5

75

S‐SCAMandMint2(see1.3.3)(Gravanteetal.,2004;Kimuraetal.,2004;Michelsetal.,2008;Brandtetal.,2009;Santoroetal.,2009;Yingetal.,2012;Cao‐Ehlkeretal.,2013; Noam et al., 2014). These modulations seemed to be irrelevant for thephysiological proliferationofneural stem cells seen in theEMX‐HCN‐573Xmice. Inagreementwiththis,theHCN3subtypeislessornotatallcAMPresponsive,anditisthemajorchannelsubtypeexpressedattheseearlyprenataltimepoints(Figure4‐1).As a result, cAMP binding would induce only a minor depolarization shift of theactivationcurve.Nevertheless,therearetwofurthermolecules,Srckinase,andPIP2,which modulate HCN channels independently of the CNBD. Although the HCN3subtypeisnotwellinvestigatedsofar,thesetwositesarewellconservedintheHCNsubtypesandaresupposed to interactwithall fourHCNsubtypes, includingHCN3.The Src kinase leads to a depolarizing shift of the activation curve upon HCNphosphorylation (Zong et al., 2005; Arinsburg et al., 2006; Li et al., 2009).Additionally,theeffectofPIP2onHCN3channelswasinvestigatedinintergeniculateleafletneurons,whereadepolarizingshiftoftheactivationcurveofapproximately17mVwasobserved(Yingetal.,2011).Thisshiftcanvaryinneuralstemandprogenitorcellsbutprobably results in ahalf‐maximal activationof approximately ‐60mV forhumanHCN3(Stieberetal.,2005)and‐78mVformouseHCN3(Mistriketal.,2005),suggesting at least a small current density in the range of the known membranepotentialsofneuralstemcells(‐40to‐70mV)(ApreaandCalegari,2012).

UsingtheHCN‐573Xmutation,doesnotexcludethatthedecreasedproliferationseeninEMX‐HCN‐DNmiceisduetotheincreasedinteractionoftheoverexpressedsubunitwith other interacting proteins. Nevertheless, in the primary cortical stem cellculture,thesamereductionintheabilitytoproliferateisseenuponIhinhibitionusingapharmacologicalblocker.Intheinvitroexperiments,noproteinwasoverexpressed,thusexcludingthepossibilityofanalteredproteininteractionsduetotheincreasedHCNproteincontentasthesinglecauseforthereducedproliferation.

In summary, the severely reduced forebrain size resulting from a decrease inproliferation in the EMX‐HCN‐DN animals are most likely due to the functionalablationofIhandnotcausedbyanoff‐targeteffectproducedbytheoverexpressionofthetransgene.ThisphenotypeispotentiallydependentontheSrckinaseand/orPIP2modulation of theHCN channel protein, but ismost likely not dependent on otherinteractingproteinsorcAMP.

5 Discussion

76

5.7 Electrophysiologicalcharacterizationofneuralstemcells

Toanalyze theexact roleof the Ih during theproliferationofneural stemcells, theelectrophysiological profile of the primary cortical stem cells were investigated invitro (Figure 4‐13). Two different macroscopic outward currents were detected,probably both potassium conducting. The current trace characteristics suggest adelayedrectifyingcurrent(IDR)andtransient,inactivatingcurrent(IA)(Figure4‐13AandB).

There is a variety of studies reporting different voltage‐gated potassium channelsexpressedinproliferativecells.Bydeterminingthemembranepotential inexcitableand non‐excitable cells the super‐family of potassium channels has an importantfunctionfortheproliferationofthesecells.Here,thefocuswillbeonneuralstemandprogenitor cells. In congruence with the presented findings, the further analyzedhuman andmurine neural stem cells exhibit two populations of outward directedpotassiumchannels,namelytheIDRandtheIA(Caietal.,2003;Lietal.,2008;Limetal., 2008; Schaarschmidt et al., 2009). There are contradictory results as to othervoltage‐gated ion channels: additional inward sodium channels in different currentdensities, a small voltage‐gated calcium current and a hyperpolarization‐activatedinwardcurrentofunknownoriginhavebeendescribed (Cai et al., 2003;Limet al.,2008). This inward current could be assigned to the HCN channels based on itsreversal potential and cesium‐dependent inhibition, yet it was also blocked uponbarium application, which is an indicator for the inwardly rectifying potassiumcurrent (IIR) (Lim et al., 2008). In the electrophysiological recordings conducted inthisproject,nohyperpolarization‐activatedortheabovementionedsodiumcurrentsweredetected,calciumcurrents,however,werenotmeasurableinthisexperimentalset up due to the addition of the calcium chelator EGTA to the pipet solution (seematerialsection).

Previous studies suggest an important role of the potassium channels on theproliferativeactivityoftheneuralstemandprogenitorcells.Inlong‐termculturesofmesencephalicprogenitorcells, thespecificblockadeof the IDRwasable toenhanceproliferationshownbyan increase inBrdU incorporation(Liebauetal.,2006).Theincrease in proliferation was also seen in experiments blocking specifically the IIRwith barium chloride, in astrocytes and adult neural progenitors in the SVZ(MacFarlaneandSontheimer,2000;Yasudaetal.,2008). In theseandotherstudiesproliferation was arrested upon the general inhibition of voltage‐gated potassiumcurrents in different systems such as human neural progenitor cells, astrocytes,oligodendrocyticprogenitorsandeveninmurineembryonalstemcells(Ghianietal.,1999;MacFarlane and Sontheimer, 2000; Yasuda et al., 2008; Schaarschmidt et al.,2009;Ngetal.,2010).

Discussion 5

77

InembryonicslicesatE12orE14IDR,calcium‐activatedpotassiumcurrents,voltage‐gatedcalciumcurrents,andinsomecellssodiumcurrentsweredetected(MienvilleandBarker,1997;PickenBahreyandMoody,2003).Thedifferentiationstatusoftheanalyzed cells was not established in these studies, leaving the possibility thatchannelexpression startswithdifferentiation, like theVGCCor sodiumchannelsasmarkerforexcitablecells(see5.5).

ThedifferenceintheelectrophysiologicalprofilesseeninthedifferentexperimentalsetupscanbeduetothevarietyofneuralprogenitorcellslikeaRG,bRG,aIPandbIP(see 1.1). The experiments were conducted at different time points during thedevelopment, and therefore with varying contributions of the different neuralprogenitors.

Besides IDR, electrophysiological recordings of embryonic stem cells of the mousedetectedIhinaminorityofinvestigatedcells(Wangetal.,2005).TheIhwasinhibitedbycesium,aknownHCNchannelblocker.NoeffectofelevatedcytosoliccAMPlevelsdue to isoproterenol, a beta‐adrenoceptor agonist were detected, suggesting thecontributionofHCN3subtype,as it isnon‐responsivetocAMP(Mistriketal.,2005;Wangetal.,2005).Inaccordancewiththis,themRNAsforHCN3and,atmuchlowerlevels,HCN2weredetected(Wangetal.,2005).Incontrast,neitherIhnorthemRNAofanysubtypeweredetectedinhumanembryonicstemcells,whileIDRwaspresent(Wangetal.,2005).Consistentwith theexperimentsconducted inneural stemandprogenitorcells,Wangetal.showedadecreaseinproliferationupontheinhibitionofvoltage‐ and calcium‐dependent potassium currents (Wang et al., 2005).Unfortunately, in this study, the inhibition of the murine Ih and its impact onproliferationwerenotinvestigated(Wangetal.,2005).

Itmustbementionedthough,thattheseexperimentswereperformedinmouseandhuman stem cell lines, which could alter the electrophysiological profile of theanalyzed cells, due to the artificial treatment and the possible changes in the cellsuponlong‐termpropagationinculture(Panetal.,2009).

These results suggest a major impact of the ion channel composition on theproliferation of neural progenitor cells in different regions and at different timepointsduringdevelopment.Manipulationof theconductanceofpotassiumchannelsas one of the major determinants of the membrane potential underlined theimportanceofthebiophysicalpropertiesoftheseearlycells.However,theseresultsdidnotshowadetectablecurrentmediatedby theHCNchannels inneuralstemorprogenitorcellsandaretherebycontrarytotheseverephenotypeoftheEMX‐HCN‐DNmiceandtheimpairedproliferationintheinvivoandinvitroexperimentsuponIhinhibition.

5 Discussion

78

The surprising result of the severe effect of the genetic and pharmacologicalinhibition of the Ih (Figure 4‐3, Figure 4‐4, Figure 4‐5, Figure 4‐7, Figure 4‐10)together with the absence of a detectable Ih in the electrophysiological recordings(Figure4‐13),ledtoseveralhypothesesregardingthecontributionofHCNchannelstotheproliferationanddifferentiationintheinvivoandinvitroexperiments.Thereare only a few reports suggesting the presence of Ih in proliferative cells. Inembryonic stemcells,Lauet al. showeda current trace, claiming tohave identifiedthe Ih,but inhibitionwithZD7288was incomplete, leaving theexact identityof thiscurrentopen(Lauetal.,2011).However,Wangetal.measuredtheIhinanembryonicstemcellline(Wangetal.,2005).

Onepossiblereasonforthisdiscrepancycouldbetheexperimentalenvironmentandits change during the electrophysiological experiment. In here presentedexperiments, the primary neural stem cells were cultured in DMEM/F‐12mediumincludingsupplements(seemethodsectionsfordetails),withabicarbonatepHbuffersystem,whileanHEPESbufferedbathsolutionwasusedintheelectrophysiologicalrecordings (seemethod section for details). This change in themedium could be apotentialstressfactorfortheneuralstemcells,whichcouldleadtoanalteredproteinsurface expression and thereby to the absence of the Ih (Cheng et al., 2016). Toanalyze and potentially avoid this problem, the primary neural stem cells could bepatchedinHEPES‐bufferedDMEMtoassurethesameenvironmenttothecells.

Furthermore, there are more physiological explanations for the absence of thecurrentwhichcouldalsobevalidforotherexperimentsconductedinacuteslicesofembryonicbrains(PickenBahreyandMoody,2002,2003).Suchpossibilitieswouldbe a small current amplitude of the instantaneous component of the Ih, or varyinglocalizationsoftheHCNsubunits.

5.7.1 TheinstantaneousIhcomponent

The Ih consists of twodifferent components: the Iins, a voltage‐independent currentandthesteady‐statecurrentIss,theslowlyactivatingcurrentuponhyperpolarization(see 1.3.1) (Macri et al., 2002; Proenza et al., 2002; Rodrigues and Oertel, 2005;Aponteetal.,2006).Inheterologoussystems,theHCN3exhibitsarelativelylargeIinswith13%ofthetotalcurrent(Mistriketal.,2005).Inaphysiologicalsystemlikethestellatecellsofthemammalianventralcochlearnucleus,theIinsexhibitssmallcurrentdensities (Rodrigues and Oertel, 2005). This small current can contribute to themembranepotentialof theneural stemcells,whichcanbedifficult tomeasure inaphysiological system, without artificial channel overexpression, if the currentamplitudesarelow.Itisknown,thattheIinscanberegulatedindependentlyoftheIss,for example by intracellular chloride concentrations, or MiRP1 co‐expression (see1.3.3) (Proenza et al., 2002; Mistrík et al., 2006). The intracellular chloride

Discussion 5

79

concentration has an inverse effect on the Iins (Mistrík et al., 2006). However, inneuralstemcells,thechlorideconcentrationisrelativelyhigh,duetotheabsenceofthe potassium/chloride co‐transporter 2 (KCC2) (Rivera et al., 1999),whichwouldlead to a potential suppressionof the Iins, arguing against the contribution of Iins inneural stem cells. The other regulatory mechanism, which could contribute to thedemonstratedeffectofHCNchannelsontheproliferationofneuralstemcells,istheinteraction with MiRP1. In CHO cells MiRP1 interaction with the HCN2 channelprotein led to an increase of Iins and a decrease in the steady state component(Proenzaetal.,2002). Incontrast, theEMX‐HCN‐573xanimals,whichomitpartsoftheC‐terminusresponsiblefortheseinteractions,showednormalbrainmorphology.This result suggests that proliferation is independent of C‐terminus interactingproteins.Nevertheless,theindependentregulationofIinsandIsssupportstheideaofthepossiblepresenceofsinglecurrentcomponent.Assumingasmallamplitudeofthevoltage‐independentIins,thiscanbedifficulttodetectinthecells,butduetoahigheropenprobability itwouldstabilize themembranepotentialof theneural stemcellsclose to the HCN reversal potential (‐20 mV to ‐40 mV) (Mistrik et al., 2005;Rodrigues and Oertel, 2005; Stieber et al., 2005) being relatively depolarized. TheinhibitionoftheIinswouldthenleadtothehyperpolarizingshift,reducingtheabilityof the cells to proliferate (Binggeli and Weinstein, 1986; Yang and Brackenbury,2013).

5.7.2 SubcellularlocalizationofHCNchannelsubunits

In the immunohistochemical investigations of our primary cortical stem cells, theHCN4 protein was localized intracellularly (Figure 4‐13 D). This localizationsuggestedashorttimespanofthemembraneintegrationoftheHCNsubunit,withasubsequentrapidinternalization.

Fromcancercelllinesitisknown,thatthemembranepotentialhyperpolarizesattheentryintheSphase,andaslowdepolarizationissuggestedattheendoftheSphaseresultinginadepolarizedpotentialintheMphase(Urregoetal.,2014).Accordingtothisscenario,theIhcouldplayamajorroleindepolarizingtheplasmamembraneandthereby facilitating the transition from S to G2/M phase. This specific role of thechannelwouldrequireashortmembraneintegrationoftheHCNchannel,followedbyafastinternalization.ThefastinternalizationoftheHCNproteinwasshownforHCN1inhippocampalpyramidalneuronsfollowingstatusepilepticusafteronehour(Junget al., 2011).This canexplain the lackofmeasurable current in ourheterogeneousneural stem cell population, containing cells in all phases of the cell cycle.Nevertheless, a short expression of the HCN channel and its Ih leading to thedepolarization of themembrane potential appears to be essential for the cell cycleprogression.Thecurrent inhibition led to thedecreasedproliferationseen in the invivoandinvitroexperiments.

5 Discussion

80

Furthermore, the immunohistochemical analysis of the primary cortical stem cells(Figure4‐13D)implieslocalizationoftheHCN4proteininthenuclearmembrane.

Differentionchannelsareexpressedinthedifferentintracellularmembraneswithavariety of functions (Checchetto et al., 2016). The nucleus is surrounded by thenuclearenvelope,abi‐layeredmembranesystemconsistingoftheinnerandtheouternuclearmembrane (Mazzantietal.,2001).Theouternuclearmembranebelongs tothe endoplasmic reticulum (ER) (Mazzanti et al., 2001). The perinuclear space isthereforeconnectedtothelumenoftheER.Thewholenuclearenvelopeisspannedby thenuclearporecomplexes,whicharehighlypermeable tosmallmoleculesandions aswell asmacromolecules (Mazzanti et al., 2001). This results in a relativelybalanced ion flux between the nucleoplasm and the cytoplasm. However, theperinuclearspacecontainshighcalciumandlowpotassiumconcentrationscomparedtothenucleoplasmorcytoplasm(Checchettoetal.,2016).Intheinnermembraneofthenuclearenvelope,differentionchannelsarelocated,e.g.ATPandcalcium‐gated,aswellasvoltage‐gatedpotassiumchannels(Checchettoetal.,2016).TheIIRchannelswere detected in cancer cell lines and rat neuronal cells. The function of thesechannelsatthenuclearlocationwasnotestablishedsofar(Stonehouseetal.,2003;Olsen and Sontheimer, 2004). Furthermore, the Kv1.3 channelwas detected in thenuclear membrane (Jang et al., 2015); its inhibition is associated with thephosphorylation and activation of cAMP response element‐binding protein (CREB)and c‐fos, two important transcription factors (Jang et al., 2015). Additionally, theEAG1orKv10.1channelwasdetectedinthenuclearmembrane,anditsphysiologicalfunctionality was determined by electrophysiological investigation of the channel(Chenet al., 2011).Theexact functionof the channelswasnot established,but thechannelproteinwasassociatedwithheterochromatin(Chenetal.,2011).Cationslikesodium,potassium,magnesiumandcalciumarelinkedtothemorecompactexistenceofDNAassociatedwithlowertranscriptionrates,suggestingonepossiblefunctionoftheionchannels(Stricketal.,2001).

ThelocalizationinthenuclearmembranecanexplaintheabsenceofthedetectableIhin primary neural stem cells and the intracellular location of the HCN4 channelprotein(Figure4‐13CandD).Furthermore,themembranepermeabilityofZD7288stillallowsinhibitionofHCNchannelslocalizedinthenuclearmembrane(HarrisandConstanti,1995). Incontrary, thewesternblotanalysisof the15daysoldembryosshowedproteinexpressionofHCN3andHCN4 in theplasmamembrane fractionofthebraintissue(Figure4‐1C).Furthermore, therestingmembranepotentialof theinnernuclearmembrane isdepolarizedatapproximately ‐30mV(Checchettoetal.,2016).AtthisdepolarizedpotentialHCNchannelswouldnotbeactive(Stieberetal.,2005;Bieletal.,2009).

Discussion 5

81

Insummary,thedatapresentedheredonotexplaintheexactmechanismastohowloss of Ih affects the proliferation of neural stem cells. The abovementionedhypothesesofasmall Ih currentdensityand/or theshortexpressionof thechannelprotein in the plasma membrane are not mutually exclusive with the nuclearlocalizationofHCNsubunits.Chenetal.showedthattheKv10.1channelisexpressedatthecellsurfacebeforeitcanbefoundintheinnernuclearmembrane,suggestingaroleinacellsignalingpathway(Chenetal.,2011).ThiscouldalsobetruefortheHCNchannels.

5.8 KnownlimitationsofZD7288

In theconductedcell cultureexperimentswithprimarycortical stemcells fromtherat, Ih was pharmacologically blocked by application of the inhibitor ZD7288. ThisinhibitorwaslongthoughttointeractspecificallywithHCNchannelsandwasusedinavarietyofelectrophysiologicalexperimentsusingconcentrationsfrom1μMto100μM(Surgesetal.,2003;Benderetal.,2005;Mominetal.,2008;Lietal.,2009;Huanget al., 2011). In heterologous expressing HEK cells, the half maximal inhibitoryconcentration(IC50)ofallfourdifferentHCNsubtypesvariedfromapproximately20to40μM(Stieberetal.,2005).

However,recentstudiesshowedunspecificinteractionswithotherionchannels.Withan IC50 of about 40 µM, ZD7288 blocked T‐type calcium channels in hippocampalneurons from the rat (Sánchez‐Alonso et al., 2008). This effect was also shown inother cell types andHEKcells,withheterologousexpressionof theCav3.1, 3.2 and3.3, thecalciumchannelsubtypesunderlying theT‐typecurrent(Felixetal.,2003).Furthermore,Wuetal.showedtheinhibitionofvoltage‐dependentsodiumcurrentsupon ZD7288 application with an IC50 of about 1.17 µM, suggesting that ZD7288blockssodiumcurrentsmoreefficientlythanHCNchannels(Wuetal.,2012).

Inthiswork,10μMor30μMZD7288wereusedinallexperiments,excepttheassayinvestigatingthedifferentiationofprimaryneuralstemcells,whereonly30μMwereapplied. These concentrations are in the range of the previously used ZD7288concentrationsandaroundthehalfmaximalinhibitoryconcentration(~20‐40μM)(Stieberetal.,2005).

Inneuralstemorprogenitorcell,most investigatorswereunable todetectvoltage‐gated‐calciumchannels(Piperetal.,2000;Lietal.,2008;Limetal.,2008),andevenwhen they did,, these currents were associated with the differentiation of theanalyzedcells(D’Ascenzoetal.,2006;Lepskietal.,2013;Tothetal.,2016).Inotherexperiments,usingmurineembryonicslices,smallcalciumcurrentsweremeasuredupondepolarization.Nevertheless,theonsetofdifferentiationinthesecellscouldnot

5 Discussion

82

beexcluded(LoTurcoetal.,1995;Caietal.,2003;PickenBahreyandMoody,2003).The closer investigation of the molecular substrate of these voltage‐gated calciumchannelsrevealedproteinsbelongingtotheL‐typefamilyofcalciumchannelswhichwerenotaffectedbytheZD7288application(Caietal.,2003;D’Ascenzoetal.,2006;Lepskietal.,2013).

Furthermore, the detection of sodium currents was inconsistent in the literatureaboutneuralstemorprogenitorcells,suggestingthattheonsetofsodiumcurrentisalsoconnectedtothecellcycleexit,andthustodifferentiationofthesecells(Piperetal., 2000; Picken Bahrey and Moody, 2003; Li et al., 2008; Lim et al., 2008). Thissuggestsapossibleroleofsodiumcurrentsinthealtereddifferentiationseenintheinvitroexperiments,butnotinthedecreasedproliferation(Figure4‐10).

Moreover, a ZD7288 induced inhibition of exocytosis was shown in neurons andpituitary lactotrophs (Chevaleyre andCastillo, 2002;Gonzalez‐Iglesias et al., 2006).Theeffectontheproliferationanddifferentiationwhichwasobservedintheprimarycortical stem cells upon application of ZD7288 presumably did not result frominhibited exocytosis, given the fact that stem cells are not especially active inexocytosis.

Inthepresentedinvestigationoftheelectrophysiologicalprofileoftheusedprimarycorticalstemcellsoftherat,calciumcurrentcouldnotbedetectedduetoacalciumchelatorinthepipettesolution.Additionally,fastsodiumcurrentwerenotdetectinthesecells.

Insummary, it isunlikelythataknownsideeffectofZD7288oncalciumorsodiumchannels inhibited the proliferation of the primary cortical stem cells used in thisstudy. This suggests that the decrease in proliferation is indeed mediated by thepharmacologicalblockadeoftheHCNchannels.Theseresultsaresupportedbytheinvivodata,usingageneticstrategytofunctionallyinactivateIh.However,itcannotbeexcluded that the temporal alterations seen in the differentiation of the primarycorticalstemcellsbytheinhibitorZD7288resultedexclusivelyfromtheinhibitionofIh. It ispossiblethatanundetected inhibitionofsodiumcurrentscontributedtotheobservedeffect.CalciumcurrentsdonotseemtobeaffectedbyZD7288,duetotheexpressionofL‐typechannels,butasanoff‐targeteffectofZD7288onT‐typecalciumchannels.However,thegeneticablationoftheIhintheEMX‐HCN‐DNmouseshowedatemporal alteration of differentiation as well, supporting the role of the ZD7288inhibitionoftheHCNchannelsasaspecificmechanismofitseffectondifferentiation.

Discussion 5

83

5.9 Conclusions

This work presents a so far unknown importance of the HCN channels and theconducted Ih for theproliferationanddifferentiationofneural stemandprogenitorcells.

A HCN subtype switchwas detectedwith HCN3 and HCN4 being highly expressedduringembryonicdevelopmentindifferentcelltypes,andHCN1andHCN2beingthemainsubtypesintheadultforebrain.TheearlyexpressionofHCN3wasdetectedindifferent neural progenitor cells while HCN4 was mainly expressed in earlyembryonicneurons.

Inatransgenicmouseline,theforebrain‐restrictedgeneticablationofIh,resultedinamicrocephaly phenotype that was caused by impaired proliferation of neuralprogenitorcells.ThisresultwasverifiedbypharmacologicalIhablationinratprimarycorticalstemcells.Asthedepolarizedrestingmembranepotentialandthemembranepotentialoscillationsduringthecellcycleareessentialforthecellcycleprogression,the proliferation rate of cortical stem cells could be decreased due to ahyperpolarizationoftherestingmembranepotentialinIh‐inhibitedcells.

In the in vivo as well as the in vitro experiments, a temporally altered pattern ofdifferentiation was detected, with the early expression of glial markers upon Ihinhibition.Onemight hypothesize that this is caused by the lengthening of the cellcycle,supportingthedifferentiationofneuralprogenitorcellsincombinationwithanimpaired L‐type calcium channel signaling in the differentiating cells due to thehyperpolarizationofthemembranepotential.

Theelectrophysiologicalinvestigationoftheprimarycorticalstemcellsdetectedtwodifferentpotassiumoutwardcurrents,theIDRandIA,whereasnoIhwasmeasurable.This couldbeexplainedbya small currentamplitudeof Ih or its component the Iinsand/or a transient presence of the HCN channel in the plasma membrane.Furthermore,theHCNchannelcanbelocalizedtothenuclearmembraneregulatingtranscriptionofgenes,forexample,byioncompositionofthenucleoplasm.

Furthermore, the HCN3 expression in human and murine cortical progenitor cellssuggests a conserved function across species.Moreover, the results obtained in ratprimarycorticalstemcellsshowing thedecreaseofproliferationupon Ih inhibition,suggest an evolutionarily conserved function of the HCN channels and theircorrespondingIh.

5 Discussion

84

In summary, this work suggests a novel role of HCN channels in the earlydevelopmental processes of the cortical structures through its function as ionchannel.

5.10 Outlook

To support the results of this study concerning the effect of the Ih ablation onproliferation, the in vivo analysis of the proliferation should be conducted atembryonicstages.ThiscouldbeachievedbyanalyzingtheEMX‐HCN‐DNtransgenicembryos at E13 or by in utero electroporation of a HCN‐DN subunit at E13, andanalyzingthepupsatE16.Thiswouldtargetthetimepointsofthepeakproliferationinthemurinecortex,andcouldaddvaluabledatatothestudy.

Furthermore, it is still not clear, how Ih influences the proliferation of neural stemcells. Analyses of awhole cell population of neural stem cells instead of individualcellstoinvestigatedifferentcellcyclestages,aswellasalargernumberofcellsinashorterperiod,wouldhelpexplaintheeffectsofIhinhibitioninneuralstemcells.Thiscould be achieved by the application of a voltage‐sensitive dye and simultaneousimaging of a population of cells. Furthermore, themembrane potential oscillationsduring the cell cycleprogressionof neural stem cells couldbe investigated via thismethod,whichwouldprovideimportantevidencetothesofarunknownmechanismofcellcycleregulationandmembranepotentialinthosecells.

Anotherwaytoaddresstheproblemofthecellcyclestatusoftheanalyzedcellsistovisualize the cell cycle state by the fluorescence ubiquitination cell cycle indicator(FUCCI) vector (Sakaue‐Sawano et al., 2008). This construct uses the oscillatingexpressionandfastdegradationofcellcycleproteins.AtruncatedversionoftheCdt1protein coupled to a red fluorophore is expressed during the G1 phase andaccumulates inG0,whiletruncatedgeminin,whichisexpressedduringS,G2andMphases, is marked with a green fluorophore (Sakaue‐Sawano et al., 2008). Thisconstructenablestheappreciationofthecellcyclephaseoflivingcellsandwouldbeusefultosortthecellspriortotheelectrophysiologicalinvestigationtoanalyzeonlycellsofonespecificcellcyclephase.PrimarycorticalstemcellstransfectedwiththeFUCCIconstructshouldshowahigherprobabilitytoconductIhduringtheSand/orG2phase,whilethegreenfluorophorecoupledgemininconstructisexpressed.

To address another hypothesis of the mechanism of the contribution of Ih to theproliferationofneuralstemandprogenitorcells,thepotentiallocalizationoftheHCNprotein in thenuclearmembraneshouldbe investigated.Thiscouldbeanalyzedbywestern blot analysis of the nuclear membrane fraction of the cells, or theimmunohistochemical investigationof theexact localizationoftheHCNchannelsby

Discussion 5

85

co‐staining with nuclear membrane markers. Furthermore, electrophysiologicalmeasurementsofthenuclearmembranecouldbeconducted.

These further investigationscouldbringadditional insights intotheroleof Ih in theproliferationofneuralstemandprogenitorcells.

6 Supplement

86

6 Supplement

6.1 ROIsofStereology

Figure6‐1RegionofInterest(ROI)forstereologicalinvestigationofsensorimotorcortexandventricularandsubventricularzone(VZ/SVZ).

A,Wild‐type(WT)mouseROIofthesensorimotorcortexandVZ/SVZwithDAPIstainingandcorrespondingbrainmapat2.31mmfrombulbusolfactorius.B,Wild‐type(WT)mouseROIforsensorimotorcortexwithDAPIstainingandcorrespondingbrainmapat3.75mmfrombulbusolfactoriusForanalysiscorticaldatafromAandBwerepooled.C,andD,EMX‐HCN‐DNmousewithROIsfromtheinvestigatedareas.Brainmapsfrom(Paxinosetal.,2007).

Supplement 6

87

6.2 VectorchartsforIUE

Figure6‐2VectorchartsoftheIUEconstructs.

A,AddgenevectorwithubiquitinousexpressedCAGpromotor(CMVenhancercombinedwithchickenbetaactinpromotor),GFPexpressionisdrivenbyaninternalribosomalentrysite(IRES)fromtheencephalomyocarditisvirus(EMCV).B,InthevectorbackboneshowninAhumanHCN‐WTorHCN‐DN(HCN‐X)wasclonedinthemultiplecloningsite.

7 References

88

7 References

AligJ,MargerL,MesircaP,EhmkeH,MangoniME,IsbrandtD(2009)ControlofheartratebycAMPsensitivityofHCNchannels.ProcNatlAcadSci106:12189–12194.

Anderson S a, Marín O, Horn C, Jennings K, Rubenstein JL (2001) Distinct corticalmigrations from the medial and lateral ganglionic eminences. Development128:353–363.

AndersonS,MioneM,YunK,RubensteinJL(1999)Differentialoriginsofneocorticalprojection and local circuit neurons: role of Dlx genes in neocorticalinterneuronogenesis.CerebCortex9:646–654.

Ang ESBC, Haydar TF, Gluncic V, Rakic P (2003) Four‐dimensional migratorycoordinates of GABAergic interneurons in the developing mouse cortex. JNeurosci23:5805–5815.

Antonopoulos J,PappasIS,Parnavelas JG(1997)Activationof theGABAAReceptorInhibits the Proliferative Effects of bFGF in Cortical Progenitor Cells. Eur JNeurosci9:291–298.

Aponte Y, Lien C‐C, Reisinger E, Jonas P (2006) Hyperpolarization‐activated cationchannels in fast‐spiking interneurons of rat hippocampus. J Physiol 574:229–243.

Aprea J, Calegari F (2012) Bioelectric State and Cell Cycle Control of MammalianNeuralStemCells.StemCellsInt2012:1–10.

Arinsburg SS, Cohen IS, Yu H‐G (2006) Constitutively Active Src Tyrosine KinaseChanges Gating of HCN4 Channels Through Direct Binding to the ChannelProteins.JCardiovascPharmacol48:1–6.

Bando Y, Hirano T, Tagawa Y (2014) Dysfunction of KCNK Potassium ChannelsImpairs Neuronal Migration in the Developing Mouse Cerebral Cortex. CerebCortex24:1017–1029.

Barbuti A, Gravante B, Riolfo M, Milanesi R, Terragni B, DiFrancesco D (2004)Localizationofpacemakerchannelsinlipidraftsregulateschannelkinetics.CircRes94:1325–1331.

BarbutiA,TerragniB,BrioschiC,DiFrancescoD(2007)Localizationoff‐channelstocaveolae mediates specific β2‐adrenergic receptor modulation of rate insinoatrialmyocytes.JMolCellCardiol42:71–78.

Battefeld A, Rocha N, Stadler K, Bräuer AU, Strauss U (2012) Distinct perinatalfeaturesofthehyperpolarization‐activatednon‐selectivecationcurrentIhintheratcorticalplate.NeuralDev7:21.

BeharTN,LiYX,TranHT,MaW,DunlapV,ScottC,BarkerJL(1996)GABAstimulates

References 7

89

chemotaxis and chemokinesis of embryonic cortical neurons via calcium‐dependentmechanisms.JNeurosci16:1808–1818.

BeharTN,SchaffneraE,ScottCa,O’ConnellC,BarkerJL(1998)Differentialresponseof cortical plate and ventricular zone cells to GABA as amigration stimulus. JNeurosci18:6378–6387.

Behar TN, Schaffner AE, Scott CA, Greene CL, Barker JL (2000) GABA receptorantagonistsmodulate postmitotic cellmigration in slice cultures of embryonicratcortex.CerebCortex10:899–909.

BeharTN,ScottCA,GreeneCL,WenX,SmithSV.,MaricD,LiuQ‐YY,ColtonCA,BarkerJL (1999)GlutamateActing atNMDAReceptors Stimulates Embryonic CorticalNeuronalMigration.JNeurosci19:4449–4461.

Ben‐Ari Y (2002) Excitatory actions of gabaduring development: the nature of thenurture.NatRevNeurosci3:728–739.

Ben‐AriY(2006)Basicdevelopmentalrulesandtheirimplicationsforepilepsyintheimmaturebrain.EpilepticDisord8:91–102.

Bender RA, Baram TZ (2008) Hyperpolarization activated cyclic‐nucleotide gated(HCN)channelsindevelopingneuronalnetworks.ProgNeurobiol86:129–140.

Bender RA, GalindoR,MameliM, Gonzalez‐vegaR, Valenzuela F, BaramTZ (2005)Synchronizednetworkactivityindevelopingrathippocampusinvolvesregionalhyperpolarization‐activatedcyclicnucleotide‐gated(HCN)channelfunction.EurJNeurosci22:2669–2674.

BenderRA,SoleymaniSV,BrewsterAL,NguyenST,BeckH,MathernGW,BaramTZ(2003) Enhanced expression of a specific hyperpolarization‐activated cyclicnucleotide‐gatedcationchannel(HCN)insurvivingdentategyrusgranulecellsofhumanandexperimentalepileptichippocampus.JNeurosci23:6826–6836.

BerridgeMJ(1995)Calciumsignallingandcellproliferation.BioEssays17:491–500.

Biel M, Wahl‐Schott C, Michalakis S, Zong X (2009) Hyperpolarization‐ActivatedCationChannels:FromGenestoFunction.PhysiolRev89:847–885.

Binggeli R, Weinstein RC (1986) Membrane potentials and sodium channels:Hypotheses for growth regulation and cancer formation based on changes insodiumchannelsandgapjunctions.JTheorBiol123:377–401.

BittmanK,OwensDF,KriegsteinaR,LoTurcoJJ(1997)Cellcouplinganduncouplingintheventricularzoneofdevelopingneocortex.JNeurosci17:7037–7044.

Blackiston DJ, McLaughlin KA, Levin M (2009) Bioelectric controls of cellproliferation: Ion channels, membrane voltage and the cell cycle. Cell Cycle29:3519–3528.

BoonstraJ,MummeryCL,TertoolenLGJ,VanDerSaagPT,DeLaatSW(1981)Cation

7 References

90

transport and growth regulation in neuroblastoma cells. Modulations of K+transportandelectricalmembranepropertiesduringthecellcycle.JCellPhysiol107:75–83.

Bordey A, Sontheimer H (1997) Postnatal development of ionic currents in rathippocampalastrocytesinsitu.JNeurophysiol78:461–477.

Borodinsky LN, Root CM, Cronin JA, Sann SB, Gu X, Spitzer NC (2004) Activity‐dependent homeostatic specification of transmitter expression in embryonicneurons.Nature429:523–530.

BortoneD,PolleuxF(2009)KCC2ExpressionPromotestheTerminationofCorticalInterneuron Migration in a Voltage‐Sensitive Calcium‐Dependent Manner.Neuron62:53–71.

Boyes J, Bolam JP, Shigemoto R, Stanford IM (2007) Functional presynaptic HCNchannelsintheratglobuspallidus.EurJNeurosci25:2081–2092.

BrandtMC,Endres‐BeckerJ,ZagidullinN,MotlochLJ,ErF,RottlaenderD,MichelsG,Herzig S, Hoppe UC (2009) Effects of KCNE2 on HCN isoforms: distinctmodulation of membrane expression and single channel properties. AJP HearCircPhysiol297:H355–H363.

BrewsterAL,ChenY,BenderRa,YehA,ShigemotoR,BaramTZ(2006)QuantitativeAnalysis and Subcellular Distribution of mRNA and Protein Expression of theHyperpolarization‐Activated Cyclic Nucleotide‐Gated Channels throughoutDevelopmentinRatHippocampus.CerebCortex17:702–712.

Brosenitsch TA, Katz DM (2002) Expression of Phox2 transcription factors andinductionof thedopaminergicphenotype inprimarysensoryneurons.MolCellNeurosci20:447–457.

BrownH,DifrancescoD(1980)Voltage‐clamp investigationsofmembranecurrentsunderlyingpace‐makeractivityinrabbitsino‐atrialnode.JPhysiol308:331–351.

Cai J, Cheng A, Luo Y, Lu C, Mattson MP, Rao MS, Furukawa K (2003) Membranepropertiesofratembryonicmultipotentneuralstemcells.JNeurochem88:212–226.

CalderondeAndaF,MadabhushiR,ReiD,MengJ,GräffJ,DurakO,MeletisK,RichterM,SchwankeB,MungenastA,TsaiL‐H(2016)Corticalneuronsgraduallyattainapost‐mitoticstate.CellRes26:1033–1047.

CalegariF,HaubensakW,HaffnerC,HuttnerWB(2005)Selectivelengtheningofthecell cycle in the neurogenic subpopulation of neural progenitor cells duringmousebraindevelopment.JNeurosci25:6533–6538.

Calegari F, Huttner WB (2003) An inhibition of cyclin‐dependent kinases thatlengthens, but does not arrest, neuroepithelial cell cycle induces prematureneurogenesis.JCellSci116:4947–4955.

References 7

91

Cao‐EhlkerX,ZongX,HammelmannV,GrunerC,FenskeS,MichalakisS,Wahl‐SchottC,BielM(2013)Up‐regulationofHyperpolarization‐activatedCyclicNucleotide‐gatedChannel3(HCN3)bySpecificInteractionwithK+ChannelTetramerizationDomain‐containingProtein3(KCTD3).JBiolChem288:7580–7589.

Capiod T (2011) Cell proliferation, calcium influx and calcium channels. Biochimie93:2075–2079.

ChanCS,GlajchKE,GertlerTS,GuzmanJN,MercerJN,LewisAS,GoldbergAB,TkatchT,ShigemotoR,FlemingSM,ChetkovichDM,OstenP,KitaH,SurmeierDJ(2011)HCNchannelopathyinexternalglobuspallidusneuronsinmodelsofParkinson’sdisease.NatNeurosci14:85–92.

Chan FK‐M, Moriwaki K, De Rosa MJ (2013) Detection of Necrosis by Release ofLactateDehydrogenaseActivity. In:Methodsinmolecularbiology(Clifton,N.J.),pp65–70.

ChandlerKJ,ChandlerRL,BroeckelmannEM,HouY,Southard‐SmithEM,MortlockDP(2007) Relevance of BAC transgene copy number in mice: transgene copynumber variation across multiple transgenic lines and correlations withtransgeneintegrityandexpression.MammGenome18:693–708.

Chattopadhyay N, Espinosa‐Jeffrey A, Tfelt‐Hansen J, Yano S, Bandyopadhyay S,Brown EM, de Vellis J (2008) Calcium receptor expression and function inoligodendrocyte commitment and lineage progression: Potential impact onreducedmyelinbasicproteininCaR‐nullmice.JNeurosciRes86:2159–2167.

Checchetto V, Teardo E, Carraretto L, Leanza L, Szabo I (2016) Physiology ofintracellular potassium channels: A unifying role as mediators of counterionfluxes?BiochimBiophysActa‐Bioenerg1857:1258–1266.

ChenJ,KriegsteinAR(2015)AGABAergicprojectionfromthezonaincertatocortexpromotescorticalneurondevelopment.Science(80‐)350:554–558.

Chen S, Wang J, Siegelbaum S a (2001) Properties of Hyperpolarization‐ActivatedPacemakerCurrentDefinedbyCoassemblyofHcn1andHcn2SubunitsandBasalModulationbyCyclicNucleotide.JGenPhysiol117:491–504.

ChenX,HartmanA,GuoS(2015)ChoosingCellFateThroughaDynamicCellCycle.CurrStemCellReports1:129–138.

ChenY,SánchezA,RubioME,KohlT,PardoLA,StühmerW(2011)FunctionalKV10.1Channels Localize to the Inner Nuclear Membrane Meuth SG, ed. PLoS One6:e19257.

Cheng Z, Teo G, Krueger S, Rock TM, Koh HW, Choi H, Vogel C (2016) Differentialdynamics of the mammalian mRNA and protein expression response tomisfoldingstress.MolSystBiol12:855–855.

Chevaleyre V, Castillo PE (2002) Assessing the role of Ih channels in synaptic

7 References

92

transmissionandmossyfiberLTP.ProcNatlAcadSci99:9538–9543.

Choi BH, LaphamLW (1978)Radial glia in the human fetal cerebrum:A combinedgolgi, immunofluorescent and electronmicroscopic study. Brain Res 148:295–311.

ChowSS,VanPetegemF,AcciliEA(2012)EnergeticsofcyclicAMPbinding toHCNchannelCterminusrevealnegativecooperativity.JBiolChem287:600–606.

ChungWK,ShinM,JaramilloTC,LeibelRL,LeDucCA,FischerSG,TzilianosE,GheithAA, Lewis AS, Chetkovich DM (2009) Absence epilepsy in apathetic, aspontaneousmutantmouselackingthehchannelsubunit,HCN2.NeurobiolDis33:499–508.

ChvatalA,BergerT,VorisekI,OrkandRK,KettenmannH,SykovaE(1997)Changesinglial K+ currents with decreased extracellular volume in developing rat whitematter.JNeurosciRes49:98–106.

Ciccolini F, Collins TJ, Sudhoelter J, Lipp P, BerridgeMJ, BootmanMD (2003) LocalandglobalspontaneouscalciumeventsregulateneuriteoutgrowthandonsetofGABAergic phenotype during neural precursor differentiation. J Neurosci23:103–111.

Cone C, Cone C (1976) Induction of mitosis in mature neurons in central nervoussystembysustaineddepolarization.Science(80‐)192:155–158.

Cone CD, Tongier M (1973) Contact inhibition of division: Involvement of theelectricaltransmembranepotential.JCellPhysiol82:373–386.

D’Ascenzo M, Piacentini R, Casalbore P, Budoni M, Pallini R, Azzena GB, Grassi C(2006) Role of L‐type Ca 2+ channels in neural stem/progenitor celldifferentiation.EurJNeurosci23:935–944.

DarzynkiewiczZ,BrunoS,DelBinoG,GorczycaW,HotzMA,LassotaP,TraganosF(1992) Features of apoptotic cells measured by flow cytometry. Cytometry13:795–808.

DeJuanRomeroC,BorrellV(2015)Coevolutionofradialglialcellsandthecerebralcortex.Glia63:1303–1319.

DeAzevedo LC, Fallet C, Moura‐Neto V, Daumas‐Duport C, Hedin‐Pereira C, Lent R(2003) Cortical radial glial cells in human fetuses: Depth‐correlatedtransformationintoastrocytes.JNeurobiol55:288–298.

DeBerg HA, Bankston JR, Rosenbaum JC, Brzovic PS, Zagotta WN, Stoll S (2015)StructuralMechanismfortheRegulationofHCNIonChannelsbytheAccessoryProteinTRIP8b.Structure23:734–744.

Demarque M, Represa A, Becq H élène, Khalilov I, Ben‐Ari Y, Aniksztejn L (2002)Paracrine Intercellular Communication by a Ca2+‐ and SNARE‐Independent

References 7

93

ReleaseofGABAandGlutamatePrior toSynapseFormation.Neuron36:1051–1061.

Desai AR, McConnell SK (2000) Progressive restriction in fate potential by neuralprogenitorsduringcerebralcorticaldevelopment.Development127:2863–2872.

Dhillon AS, Hagan S, Rath O, Kolch W (2007) MAP kinase signalling pathways incancer.Oncogene26:3279–3290.

Dibbens LM, Reid CA, Hodgson B, Thomas EA, Phillips AM, Gazina E, Cromer BA,Clarke AL, Baram TZ, Scheffer IE, Berkovic SF, Petrou S (2010) Augmentedcurrents of an HCN2 variant in patients with febrile seizure syndromes. AnnNeurol67:542–546.

DoetschF(2003)Theglialidentityofneuralstemcells.NatNeurosci6:1127–1134.

Doetsch F, Caillé I, Lim DA, García‐Verdugo JM, Alvarez‐Buylla A (1999)SubventricularZoneAstrocytesAreNeuralStemCells in theAdultMammalianBrain.Cell97:703–716.

Dyhrfjeld‐Johnsen J, Morgan RJ, Soltesz I (2009) Double Trouble? Potential forHyperexcitabilityFollowingBothChannelopathicup‐andDownregulationofI(h)inEpilepsy.FrontNeurosci3:25–33.

Emery EC, Young GT, Berrocoso EM, Chen L, McNaughton PA (2011) HCN2 IonChannelsPlayaCentralRoleinInflammatoryandNeuropathicPain.Science(80‐)333:1462–1467.

Er F, Larbig R, Ludwig A, Biel M, Hofmann F, Beuckelmann DJ, Hoppe UC (2003)Dominant‐negative suppression of HCN channels markedly reduces the nativepacemaker current I(f) and undermines spontaneous beating of neonatalcardiomyocytes.Circulation107:485–489.

Eszterhas SK, Bouhassira EE, Martin DIK, Fiering S (2002) Transcriptionalinterference by independently regulated genes occurs in any relativearrangementofthegenesandisinfluencedbychromosomalintegrationposition.MolCellBiol22:469–479.

Feldman JD, Vician L, Crispino M, Tocco G, Marcheselli VL, Bazan NG, Baudry M,Herschman HR (1998) KID‐1, a Protein Kinase Induced by Depolarization inBrain.JBiolChem273:16535–16543.

FelixR,SandovalA,SánchezD,GómoraJC,Vega‐BeltránJLDla,TreviñoCL,DarszonA(2003) ZD7288 inhibits low‐threshold Ca2+ channel activity and regulatesspermfunction.BiochemBiophysResCommun311:187–192.

FenskeS,KrauseSC,HassanSIH,BecirovicE,AuerF,BernardR,KupattC,LangeP,ZieglerT,WotjakCT,ZhangH,HammelmannV,PaparizosC,BielM,Wahl‐SchottCA (2013)Sick sinus syndrome inHCN1‐Deficientmice.Circulation128:2585–2594.

7 References

94

Fenske S, Mader R, Scharr A, Paparizos C, Cao‐Ehlker X, Michalakis S, Shaltiel L,WeidingerM,Stieber J,Feil S,FeilR,HofmannF,Wahl‐SchottC,BielM(2011)HCN3Contributes to theVentricularActionPotentialWaveform in theMurineHeart.CircRes109:1015–1023.

FerryS,TraiffortE,StinnakreJ,RuatM(2000)Developmentalandadultexpressionofratcalcium‐sensingreceptortranscriptsinneuronsandoligodendrocytes.EurJNeurosci12:872–884.

FishellG,RudyB(2011)MechanismsofInhibitionwithintheTelencephalon:“WheretheWildThingsAre.”AnnuRevNeurosci34:535–567.

FlorioMet al. (2015)Human‐specificgeneARHGAP11Bpromotesbasalprogenitoramplificationandneocortexexpansion.Science(80‐)347:1465–1470.

Frace AM,Maruoka F, Noma A (1992) Current By External Anions in Rabbit Sino‐AtrialNode.JPhysiol453,:307–318.

Francis F, Meyer G, Fallet‐Bianco C, Moreno S, Kappeler C, Socorro AC, Tuy FPD,BeldjordC,ChellyJ(2006)Humandisordersofcorticaldevelopment:frompasttopresent.EurJNeurosci23:877–893.

Franklin KBJ, Paxinos G (2007) TheMouse Brain in Stereotaxic Coordinates, ThirdEdit.

Gao P, Postiglione MP, Krieger TG, Hernandez L, Wang C, Han Z, Streicher C,Papusheva E, Insolera R, Chugh K, Kodish O, Huang K, Simons BD, Luo L,HippenmeyerS, Shi S‐H (2014)DeterministicProgenitorBehaviorandUnitaryProductionofNeuronsintheNeocortex.Cell159:775–788.

GaussR,SeifertR,KauppUB(1998)Molecularidentificationofahyperpolarization‐activatedchannelinseaurchinsperm.Nature393:583–587.

GeW‐P,MiyawakiA,GageFH,JanYN,JanLY(2012)Localgenerationofgliaisamajorastrocytesourceinpostnatalcortex.Nature484:376–380.

GerberAM,Beaman‐HallCM,MathurA,VallanoMLou(2010)Reducedblockadebyextracellular Mg2+ is permissive to NMDA receptor activation in cerebellargranuleneuronsthatmodelamigratoryphenotype.JNeurochem114:191–202.

GerdesJ,LiL,SchlueterC,DuchrowM,WohlenbergC,GerlachC,StahmerI,KlothS,Brandt E, Flad HD (1991) Immunobiochemical and molecular biologiccharacterization of the cell proliferation‐associated nuclear antigen that isdefinedbymonoclonalantibodyKi‐67.AmJPathol138:867–873.

GhianiCa,YuanX,Eisen aM,KnutsonPL,DePinhoRa,McBainCJ,GalloV (1999)Voltage‐activated K+ channels and membrane depolarization regulateaccumulationofthecyclin‐dependentkinaseinhibitorsp27(Kip1)andp21(CIP1)inglialprogenitorcells.JNeurosci19:5380–5392.

References 7

95

Ghosh A, Antonini A, McConnell SK, Shatz CJ (1990) Requirement for subplateneuronsintheformationofthalamocorticalconnections.Nature347:179–181.

GilV,NocentiniS,delRÃ‐oJA(2014)HistoricalfirstdescriptionsofCajalâ€“Retziuscells:frompioneerstudiestocurrentknowledge.FrontNeuroanat8:32.

GoebbelsS,BormuthI,BodeU,HermansonO,SchwabMH,NaveK‐A(2006)Genetictargetingofprincipalneurons inneocortexandhippocampusofNEX‐Cremice.genesis44:611–621.

Gonzalez‐Iglesias AE, Kretschmannova K, Tomic M, Stojilkovic SS (2006) ZD7288inhibitsexocytosisinanHCN‐independentmanneranddownstreamofvoltage‐gated calcium influx in pituitary lactotrophs. Biochem Biophys Res Commun346:845–850.

Gorski J a, Talley T, Qiu M, Puelles L, Rubenstein JLR, Jones KR (2002) Corticalexcitatory neurons and glia, but not GABAergic neurons, are produced in theEmx1‐expressinglineage.JNeurosci22:6309–6314.

Gravante B, Barbuti A, Milanesi R, Zappi I, Viscomi C, DiFrancesco D (2004)Interaction of the pacemaker channel HCN1 with filamin A. J Biol Chem279:43847–43853.

GrayLS, SchiffD,MacdonaldTL (2013)Amodel for the regulationofT‐typeCa2+channels inproliferation:roles instemcellsandcancer.ExpertRevAnticancerTher13:589–595.

Gu X, Spitzer NC (1995) Distinct aspects of neuronal differentiation encoded byfrequencyofspontaneousCa2+transients.Nature375:784–787.

Habela CW,OlsenML, SontheimerH (2008) ClC3 Is a Critical Regulator of the CellCycleinNormalandMalignantGlialCells.JNeurosci28:9205–9217.

HallowsJL,TempelBL(1998)ExpressionofKv1.1,aShaker‐likepotassiumchannel,is temporally regulated in embryonic neurons and glia. J Neurosci 18:5682–5691.

Hanson MG, Landmesser LT (2004) Normal patterns of spontaneous activity arerequiredforcorrectmotoraxonguidanceandtheexpressionofspecificguidancemolecules.Neuron43:687–701.

Harris NC, Constanti A (1995) Mechanism of block by ZD 7288 of thehyperpolarization‐activated inward rectifying current in guinea pig substantianigraneuronsinvitro.JNeurophysiol74:2366–2378.

HarzheimD,PfeifferKH,FabritzL,KremmerE,BuchT,WaismanA,KirchhofP,KauppUB, Seifert R (2008) Cardiac pacemaker function of HCN4 channels inmice isconfined to embryonic development and requires cyclicAMP. EMBO J 27:692–703.

7 References

96

HaubensakW, Attardo A, DenkW, HuttnerWB (2004) Neurons arise in the basalneuroepithelium of the early mammalian telencephalon: a major site ofneurogenesis.Pnas101:3196–3201.

Haydar TF, Wang F, Schwartz ML, Rakic P (2000) Differential Modulation ofProliferationintheNeocorticalVentricularandSubventricularZones.JNeurosci125:2621–2629.

HeckN, KilbW, Reiprich P, KubotaH, FurukawaT, FukudaA, LuhmannHJ (2007)GABA‐Areceptors regulateneocorticalneuronalmigration invitroand invivo.CerebCortex17:138–148.

HegleAP,MarbleDD,WilsonGF(2006)Avoltage‐drivenswitchforion‐independentsignalingbyether‐a‐go‐goK+channels.ProcNatlAcadSci103:2886–2891.

HerrmannS,SchnorrS,LudwigA(2015)HCNChannels—ModulatorsofCardiacandNeuronalExcitability.IntJMolSci16:1429–1447.

Herrmann S, Stieber J, Stöckl G, Hofmann F, Ludwig A (2007) HCN4 provides a“depolarizationreserve”andisnotrequiredforheartrateaccelerationinmice.EMBOJ26:4423–4432.

HerschmanHR,FergusonGD,FeldmanJD,Farias‐EisnerR,VicianL(2000)Searchingfor depolarization‐induced genes that modulate synaptic plasticity andneurotrophin‐induced genes thatmediate neuronal differentiation.NeurochemRes25:591–602.

Heuermann RJ, Jaramillo TC, Ying S‐W, Suter B a., Lyman K a., Han Y, Lewis AS,HamptonTG,ShepherdGMG,GoldsteinPa.,ChetkovichDM(2016)ReductionofthalamicandcorticalIhbydeletionofTRIP8bproducesamousemodelofhumanabsenceepilepsy.NeurobiolDis85:81–92.

Hicks MR, Damianoglou A, Rodger A, Dafforn TR (2008) Folding and MembraneInsertionofthePore‐FormingPeptideGramicidinOccurasaConcertedProcess.JMolBiol383:358–366.

HigashimoriH,SontheimerH(2007)RoleofKir4.1channelsingrowthcontrolofglia.Glia55:1668–1679.

HiraiK,YoshiokaH,KiharaM,HasegawaK,SakamotoT,SawadaT,FushikiS(1999)InhibitingneuronalmigrationbyblockingNMDAreceptorsintheembryonicratcerebralcortex:Atissueculturestudy.DevBrainRes114:63–67.

Hoerder‐SuabedissenA,Molnár Z (2015)Development, evolution and pathology ofneocorticalsubplateneurons.NatRevNeurosci16:133–146.

HoffmannEK,LambertIH,PedersenSF(2009)PhysiologyofCellVolumeRegulationinVertebrates.PhysiolRev89:193–277.

HonarnejadK,HermsJ(2012)Presenilins:Roleincalciumhomeostasis.IntJBiochem

References 7

97

CellBiol44:1983–1986.

Hu L, SantoroB, SaponaroA, LiuH,Moroni A, SiegelbaumS (2013)Binding of theauxiliary subunitTRIP8b toHCNchannels shifts themodeof actionof cAMP. JGenPhysiol142:599–612.

Huang Z, Lujan R, Kadurin I, Uebele VN, Renger JJ, Dolphin AC, Shah MM (2011)Presynaptic HCN1 channels regulate Cav3.2 activity and neurotransmission atselectcorticalsynapses.NatNeurosci14:478–486.

Huang Z,WalkerMC, ShahMM (2009) Loss of Dendritic HCN1 Subunits EnhancesCorticalExcitabilityandEpileptogenesis.JNeurosci29:10979–10988.

Inoue K, Furukawa T, Kumada T, Yamada J, Wang T, Inoue R, Fukuda A (2012)Taurine inhibits K +‐Cl ‐ cotransporter KCC2 to regulate embryonic Cl ‐homeostasis viawith‐no‐lysine (WNK)proteinkinase signalingpathway. JBiolChem287:20839–20850.

IwasatoT,Datwani a,Wolf aM,NishiyamaH, Taguchi Y, Tonegawa S, Knöpfel T,ErzurumluRS,ItoharaS(2000)Cortex‐restricteddisruptionofNMDAR1impairsneuronalpatternsinthebarrelcortex.Nature406:726–731.

JangSH,ByunJK,JeonW,ChoiSY,ParkJ,LeeBH,YangJE,ParkJB,O’GradySM,KimD,RyuPD,JooS,LeeSY(2015)NuclearLocalizationandFunctionalCharacteristicsofVoltage‐gatedPotassiumChannelKv1.3.JBiolChem290:12547–12557.

JovicD,Sakaue‐SawanoA,AbeT,ChoC‐S,NagaokaM,MiyawakiA,AkaikeT(2013)DirectobservationofcellcycleprogressioninlivingmouseembryonicstemcellsonanextracellularmatrixofE‐cadherin.Springerplus2:585.

JungS,JonesTD,LugoJN,SheerinAH,MillerJW,D’AmbrosioR,AndersonAE,PoolosNP (2007)Progressive dendriticHCN channelopathyduring epileptogenesis intheratpilocarpinemodelofepilepsy.JNeurosci27:13012–13021.

JungS,WarnerLN,PitschJ,BeckerAJ,PoolosNP(2011)RapidLossofDendriticHCNChannel Expression in Hippocampal Pyramidal Neurons following StatusEpilepticus.JNeurosci31:14291–14295.

Kalil K, Li L, Hutchins BI (2011) Signaling Mechanisms in Cortical Axon Growth,Guidance,andBranching.FrontNeuroanat5:1–15.

KanyshkovaT,MeuthP, Bista P, Liu Z, EhlingP,DoengiM, ChetkovichDM, PapeH(2012) Differential regulation of HCN channel isoform expression in thalamicneuronsofepilepticandnon‐epilepticratstrains.NeurobiolDis45:450–461.

KanyshkovaT,PawlowskiM,MeuthP,DubeC,BenderRa,BrewsterAL,BaumannA,Baram TZ, Pape H‐C, Budde T (2009) Postnatal Expression Pattern of HCNChannel Isoforms in Thalamic Neurons: Relationship to Maturation ofThalamocorticalOscillations.JNeurosci29:8847–8857.

7 References

98

Katz LC, ShatzCJ (1996) SynapticActivity and theConstructionofCortical circuits.Science274:1133–1138.

KauppUB,SeifertR(2001)MolecularDiversityofPacemakerIonChannels.AnnuRevPhysiol63:235–257.

Kessaris N, Fogarty M, Iannarelli P, Grist M, Wegner M, Richardson WD (2006)Competingwavesofoligodendrocytesintheforebrainandpostnataleliminationofanembryoniclineage.NatNeurosci9:173–179.

KhazipovR,LuhmannHJ(2006)Earlypatternsofelectricalactivityinthedevelopingcerebralcortexofhumansandrodents.TrendsNeurosci29:414–418.

KilbW,LuhmannHJ(2000)Characterizationofahyperpolarization‐activatedinwardcurrentinCajal‐Retziuscellsinratneonatalneocortex.JNeurophysiol84:1681–1691.

Kimura K, Kitano J, Nakajima Y, Nakanishi S (2004) Hyperpolarization‐activated,cyclic nucleotide‐gated HCN2 cation channel forms a protein assembly withmultiple neuronal scaffold proteins in distinct modes of protein‐proteininteraction.GenestoCells9:631–640.

Knop GC, Seeliger MW, Thiel F, Mataruga A, Kaupp UB, Friedburg C, Tanimoto N,Müller F (2008) Light responses in the mouse retina are prolonged upontargeteddeletionoftheHCN1channelgene.EurJNeurosci28:2221–2230.

Komuro H, Rakic P (1996) Intracellular Ca2+ fluctuations modulate the rate ofneuronalmigration.Neuron17:275–285.

Kressin K, Kuprijanova E, Jabs R, Seifert G, Steinhäuser C (1995) DevelopmentalregulationofNa+andK+conductancesinglialcellsofmousehippocampalbrainslices.Glia15:173–187.

KriegsteinA,Alvarez‐BuyllaA(2009)Theglialnatureofembryonicandadultneuralstemcells.AnnuRevNeurosci32:149–184.

KriegsteinAR,GoetzM(2003)Radialgliadiversity:Amatterofcellfate.Glia43:37–43.

Kriegstein AR, Noctor SC (2004) Patterns of neuronal migration in the embryoniccortex.TrendsNeurosci27:392–399.

Kwan KY, Sestan N, Anton ES (2012) Transcriptional co‐regulation of neuronalmigrationandlaminaridentityintheneocortex.Development139:1535–1546.

LangF,FöllerM,LangK,LangP,RitterM,VereninovA,SzaboI,HuberSM,GulbinsE(2007)CellVolumeRegulatoryIonChannelsinCellProliferationandCellDeath.In:MethodsinEnzymology,pp209–225.

LauY‐T,WongC‐K,LuoJ,LeungL‐H,TsangP‐F,BianZ‐X,TsangS‐Y(2011)Effectsofhyperpolarization‐activated cyclic nucleotide‐gated (HCN) channel blockers on

References 7

99

the proliferation and cell cycle progression of embryonic stem cells. PflugersArch461:191–202.

LaursenM, Yatime L, Nissen P, Fedosova NU (2013) Crystal structure of the high‐affinityNa+,K+‐ATPase‐ouabaincomplexwithMg2+boundinthecationbindingsite.ProcNatlAcadSci110:10958–10963.

Lavdas AA, Grigoriou M, Pachnis V, Parnavelas JG (1999) The medial ganglioniceminencegivesrisetoapopulationofearlyneuronsinthedevelopingcerebralcortex.JNeurosci19:7881–7888.

Leclerc C,Néant I,MoreauM (2012) The calcium: an early signal that initiates theformationofthenervoussystemduringembryogenesis.FrontMolNeurosci5:3.

Lepski G, Jannes CE, Nikkhah G, Bischofberger J (2013) cAMP promotes thedifferentiationofneuralprogenitorcellsinvitroviamodulationofvoltage‐gatedcalciumchannels.FrontCellNeurosci7:155.

Letinic K, Zoncu R, Rakic P (2002) Origin of GABAergic neurons in the humanneocortex.Nature417:645–649.

LevisonSW,GoldmanJE(1993)Botholigodendrocytesandastrocytesdevelopfromprogenitors in the subventricular zone of postnatal rat forebrain. Neuron10:201–212.

LewisAS,ChetkovichDM(2011)HCNchannelsinbehaviorandneurologicaldisease:Toohyperornotactiveenough?MolCellNeurosci46:357–367.

LewisAS,SchwartzE,ChanCS,NoamY,ShinM,WadmanWJ,SurmeierDJ,BaramTZ,MacdonaldRL,ChetkovichDM(2009)Alternativelyspliced isoformsofTRIP8bdifferentially control h channel trafficking and function. J Neurosci 29:6250–6265.

LewisAS,VaidyaSP,BlaissCa,LiuZ,StoubTR,BragerDH,ChenX,BenderRa,EstepCM,PopovAB,KangCE,VanVeldhovenPP,BaylissDa,NicholsonDa,PowellCM,JohnstonD, ChetkovichDM (2011)Deletion of the hyperpolarization‐activatedcyclicnucleotide‐gatedchannelauxiliarysubunitTRIP8bimpairshippocampalIhlocalization and function and promotes antidepressant behavior in mice. JNeurosci31:7424–7440.

LiC‐H,ZhangQ,TengB,Mustafa SJ,Huang J‐Y,YuH‐G (2009)SrcTyrosineKinaseAltersGatingofHyperpolarization‐ActivatedHCN4PacemakerChannelthroughTyr531.AmJPhysiol294:1–20.

Li T, Jiang L, Chen H, Zhang X (2008) Characterization of Excitability and Voltage‐gated Ion Channels of Neural Progenitor Cells in Rat Hippocampus. J MolNeurosci35:289–295.

LiebauS,PröpperC,BöckersT,Lehmann‐HornF,StorchA,GrissmerS,WittekindtOH(2006) Selective blockage of K v 1.3 and K v 3.1 channels increases neural

7 References

100

progenitorcellproliferation.JNeurochem99:426–437.

Lim CG, Kim S‐S, Suh‐Kim H, Lee Y‐D, Ahn SC (2008) Characterization of IonicCurrentsinHumanNeuralStemCells.KoreanJPhysiolPharmacol12:131.

LinJjH‐C,TakanoT,ArcuinoG,WangX,HuF,DarzynkiewiczZ,NunesM,GoldmanSA,Nedergaard M (2007) Purinergic signaling regulates neural progenitor cellexpansionandneurogenesis.DevBiol27:380–392.

LinYC,HuangJ,ZhangQ,HollanderJM,FrisbeeJC,MartinKH,NestorC,GoodmanR,Yu HG (2010) Inactivation of L‐type calcium channel modulated by HCN2channel.AJPCellPhysiol298:C1029–C1037.

LiodisP,DenaxaM,GrigoriouM,Akufo‐AddoC,YanagawaY,PachnisV(2007)Lhx6Activity Is Required for the Normal Migration and Specification of CorticalInterneuronSubtypes.JNeurosci27:3078–3089.

Lörincz A, Notomi T, Tamás G, Shigemoto R, Nusser Z (2002) Polarized andcompartment‐dependent distribution of HCN1 in pyramidal cell dendrites. NatNeurosci5:1185–1193.

LoTurco JJ, Owens DF, Heath MJS, Davis MBE, Kriegstein AR (1995) GABA andglutamatedepolarizecorticalprogenitorcellsandinhibitDNAsynthesis.Neuron15:1287–1298.

LudwigA,BuddeT,StieberJ,MoosmangS,WahlC,HolthoffK,LangebartelsA,WotjakC,MunschT,ZongX,FeilS,FeilR,LancelM,ChienKR,KonnerthA,PapeH‐C,BielM,HofmannF (2003)Absenceepilepsyandsinusdysrhythmia inmice lackingthepacemakerchannelHCN2.EMBOJ22:216–224.

Ludwig A, Zong X, Jeglitsch M, Hofmann F, Biel M (1998) A family ofhyperpolarization‐activatedmammaliancationchannels.Nature393:587–591.

Ludwig A, Zong X, Stieber J, Hullin R, Hofmann F, Biel M (1999) Two pacemakerchannelsfromhumanheartwithprofoundlydifferentactivationkinetics.EMBOJ18:2323–2329.

Luján R, Albasanz JL, Shigemoto R, Juiz JM (2005) Preferential localization of thehyperpolarization‐activatedcyclicnucleotide‐gatedcationchannelsubunitHCN1inbasketcellterminalsoftheratcerebellum.EurJNeurosci21:2073–2082.

Lukaszewicz A, Savatier P, Cortay V, Giroud P, Huissoud C, BerlandM, KennedyH,Dehay C (2005) G1 Phase Regulation, Area‐Specific Cell Cycle Control, andCytoarchitectonicsinthePrimateCortex.Neuron47:353–364.

LukaszewiczA,SavatierP,CortayV,KennedyH,DehayC(2002)Contrastingeffectsofbasicfibroblastgrowthfactorandneurotrophin3oncellcyclekineticsofmousecorticalstemcells.JNeurosci22:6610–6622.

Lupica CR, Bell J a, Hoffman a F, Watson PL (2001) Contribution of the

References 7

101

hyperpolarization‐activated current (I(h)) to membrane potential and GABAreleaseinhippocampalinterneurons.JNeurophysiol86:261–268.

Ma Y, Hu H, Berrebi AS, Mathers PH, Agmon A (2006) Distinct Subtypes ofSomatostatin‐ContainingNeocorticalInterneuronsRevealedinTransgenicMice.JNeurosci26:5069–5082.

MacFarlane SN, SontheimerH (2000)ModulationofKv1.5 currentsby Src tyrosinephosphorylation: potential role in the differentiation of astrocytes. J Neurosci20:5245–5253.

Macri V, Angoli D, Accili E a (2012) Architecture of the HCN selectivity filter andcontrolofcationpermeation.SciRep2:894.

Macri V, Proenza C, Agranovich E, Angoli D, Accili EA (2002) Separable gatingmechanismsinamammalianpacemakerchannel.JBiolChem277:35939–35946.

Magee JC (1999)Dendritic lh normalizes temporal summation in hippocampal CA1neurons.NatNeurosci2:508–514.

ManentJ‐B,DemarqueM,JorqueraI,PellegrinoC,Ben‐AriY,AniksztejnL,RepresaA(2005) A Noncanonical Release of GABA and Glutamate Modulates NeuronalMigration.JNeurosci25:4755–4765.

Manent J‐B, Represa A (2007) Neurotransmitters and brain maturation: earlyparacrine actions of GABA and glutamate modulate neuronal migration.Neuroscientist13:268–279.

Marín O, ValienteM, Ge X, Tsai L‐H (2010) Guiding neuronal cell migrations. ColdSpringHarbPerspectBiol2:a001834.

Mazzanti M, Bustamante JO, Oberleithner H (2001) Electrical dimension of thenuclearenvelope.PhysiolRev81:1–19.

McCormick DA, Pape HC (1990) Properties of a hyperpolarization‐activated cationcurrentanditsroleinrhythmicoscillationinthalamicrelayneurones.JPhysiol431:291–318.

MerseburgA(2016)HCN/h‐channeldeficiency in forebrainneurons impairsearlypostnatal development and alters neuronal network activity in mice ( Musmusculus,Linnaeus1758).Dissertation,UniversitätHamburg.

MesircaPetal.(2014)Cardiacarrhythmiainducedbygeneticsilencingof“funny”(f)channelsisrescuedbyGIRK4inactivation.NatCommun5:4664.

MichelsG,ErF,KhanIF,Endres‐BeckerJ,BrandtMC,GassanowN,JohnsDC,HoppeUC (2008)K+channel regulatorKCR1suppressesheart rhythmbymodulatingthepacemakercurrentIf.PLoSOne3.

Mienville JM, Barker JL (1997) Potassium current expression during prenatalcorticogenesisintherat.Neuroscience81:163–172.

7 References

102

MistrikP,MaderR,MichalakisS,WeidingerM,PfeiferA,BielM(2005)TheMurineHCN3 Gene Encodes a Hyperpolarization‐activated Cation Channel with SlowKinetics and Unique Response to Cyclic Nucleotides. J Biol Chem 280:27056–27061.

MistríkP,PfeiferA,BielM (2006)TheenhancementofHCNchannel instantaneouscurrent facilitated by slow deactivation is regulated by intracellular chlorideconcentration.PflugersArchEurJPhysiol452:718–727.

Molnár Z (2011) Evolution of Cerebral Cortical Development. Brain Behav Evol78:94–107.

Molyneaux BJ, Arlotta P, Menezes JRL, Macklis JD (2007) Neuronal subtypespecificationinthecerebralcortex.NatRevNeurosci8:427–437.

MominA,CadiouH,MasonA,McNaughtonPa.(2008)Roleofthehyperpolarization‐activatedcurrentIhinsomatosensoryneurons.JPhysiol586:5911–5929.

MoodyWJ,BosmaMM(2005) IonChannelDevelopment,SpontaneousActivity,andActivity‐DependentDevelopmentinNerveandMuscleCells.PhysiolRev85:883–941.

Moosmang S, BielM,HofmannF, Ludwig a (1999)Differential distribution of fourhyperpolarization‐activatedcationchannelsinmousebrain.BiolChem380:975–980.

MuchB,Wahl‐SchottC,ZongX,SchneiderA,BaumannL,MoosmangS,LudwigA,BielM (2003) Role of SubunitHeteromerization andN‐LinkedGlycosylation in theFormation of Functional Hyperpolarization‐activated Cyclic Nucleotide‐gatedChannels.JBiolChem278:43781–43786.

Munsch T, Pape HC (1999) Upregulation of the hyperpolarization‐activated cationcurrentinratthalamicrelayneuronesbyacetazolamide.JPhysiol519Pt2:505–514.

NadarajahB,BrunstromJE,GrutzendlerJ,WongRO,PearlmanaL(2001)Twomodesof radial migration in early development of the cerebral cortex. Nat Neurosci4:143–150.

Nadarajah B, Parnavelas JG (2002)Modes of neuronalmigration in the developingcerebralcortex.NatRevNeurosci3:423–432.

Nava C et al. (2014) De novo mutations in HCN1 cause early infantile epilepticencephalopathy.NatGenet.

Ng S‐Y, Chin C‐H, Lau Y‐T, Luo J, Wong C‐K, Bian Z‐X, Tsang S‐Y (2010) Role ofvoltage‐gatedpotassiumchannels in the fatedeterminationof embryonic stemcells.JCellPhysiol224:165–177.

NoamY,EhrengruberMU,KohA,FeyenP,MandersEMM,AbbottGW,WadmanWJ,

References 7

103

Baram TZ (2014) Filamin A promotes dynamin‐dependent internalization ofhyperpolarization‐activatedcyclicnucleotide‐gatedtype1(HCN1)channelsandrestrictsIhinhippocampalneurons.JBiolChem289:5889–5903.

NoctorSC,Martínez‐CerdeñoV,IvicL,KriegsteinAR(2004)Corticalneuronsariseinsymmetricandasymmetricdivisionzonesandmigratethroughspecificphases.NatNeurosci7:136–144.

NolanMF,MalleretG,LeeKH,GibbsE,DudmanJT,SantoroB,YinD,ThompsonRF,SiegelbaumS a,KandelER,MorozovA (2003)Thehyperpolarization‐activatedHCN1 channel is important for motor learning and neuronal integration bycerebellarPurkinjecells.Cell115:551–564.

Notomi T, Shigemoto R (2004) Immunohistochemical localization of Ih channelsubunits,HCN1‐4,intheratbrain.JCompNeurol471:241–276.

Nowakowski RS, Lewin SB, Miller MW (1989) BromodeoxyuridineimmunohistochemicaldeterminationofthelengthsofthecellcycleandtheDNA‐synthetic phase for an anatomically defined population. J Neurocytol 18:311–318.

OlsenML,SontheimerH(2004)MislocalizationofKirchannelsinmalignantglia.Glia46:63–73.

OmelyanenkoA,SekyrovaP,AndängM(2016)ZD7288,ablockeroftheHCNchannelfamily, increases doubling time ofmouse embryonic stem cells andmodulatesdifferentiationoutcomesinacontext‐dependentmanner.Springerplus5:41.

Owens DF, Kriegstein AR (1998) Patterns of intracellular calcium fluctuation inprecursorcellsoftheneocorticalventricularzone.JNeurosci18:5374–5388.

Pan C, Kumar C, Bohl S, Klingmueller U, Mann M (2009) Comparative ProteomicPhenotypingofCellLinesandPrimaryCellstoAssessPreservationofCellType‐specificFunctions.MolCellProteomics8:443–450.

PangT,AtefyR,SheenV(2008)Malformationsofcorticaldevelopment.Neurologist14:141–164.

Pardo LA, del Camino D, Sánchez A, Alves F, Brüggemann A, Beckh S, StühmerW(1999)OncogenicpotentialofEAGK(+)channels.EMBOJ18:5540–5547.

Pardo LA, Stuhmer W (2008) Eag1: An Emerging Oncological Target. Cancer Res68:1611–1613.

Paxinos G, Hallyday G,Watson C, Koutcherov Y,WangHH,Halliday GM,Watson C,KoutcherovY,WangHH(2007)AtlasoftheDevelopingMouseBrainatE17.5,P0andP6,FirstEdit.AcademicPress.

Pian P, Bucchi A, Robinson RB, Siegelbaum S a (2006) Regulation of gating andrundown of HCN hyperpolarization‐activated channels by exogenous and

7 References

104

endogenousPIP2.JGenPhysiol128:593–604.

Picken Bahrey HL, Moody WJ (2002) Early Development of Voltage‐Gated IonCurrents and Firing Properties in Neurons of the Mouse Cerebral Cortex. JNeurophysiol89:1761–1773.

Picken Bahrey HL, Moody WJ (2003) Voltage‐gated currents, dye and electricalcouplingintheembryonicmouseneocortex.CerebCortex13:239–251.

Piper DR, Mujtaba T, Rao MS, Lucero MT (2000) Immunocytochemical andphysiological characterization of a population of cultured human neuralprecursors.JNeurophysiol84:534–548.

PlatoshynO, Golovina VA, Bailey CL, LimsuwanA, Krick S, JuhaszovaM, Seiden JE,RubinLJ,Yuan JX‐J (2000)Sustainedmembranedepolarizationandpulmonaryarterysmoothmusclecellproliferation.AmJPhysiolCellPhysiol279:C1540‐9.

Proenza C, Angoli D, Agranovich E, Macri V, Accili EA (2002) Pacemaker channelsproduceaninstantaneouscurrent.JBiolChem277:5101–5109.

ProenzaC,YellenG(2006)DistinctPopulationsofHCNPacemakerChannelsProduceVoltage‐dependent and Voltage‐independent Currents. J Gen Physiol 127:183–190.

Rakic P (1972) Mode of cell migration to the superficial layers of fetal monkeyneocortex.JCompNeurol145:61–83.

Rakic P (1974) Neurons in Rhesus Monkey Visual Cortex: Systematic RelationbetweenTimeofOriginandEventualDisposition.Science(80‐)183:425–427.

RakicP(2009)Evolutionoftheneocortex:aperspectivefromdevelopmentalbiology.NatRevNeurosci10:724–735.

ReiprichP,KilbW,LuhmannHJ (2005)NeonatalNMDAreceptorblockadedisturbsneuronalmigration in rat somatosensory cortex in vivo. CerebCortex 15:349–358.

ResendeRR,AdhikariA,daCostaJL,Loren??onE,LadeiraMS,GuatimosimS,KiharaAH, Ladeira LO (2010) Influence of spontaneous calcium events on cell‐cycleprogressioninembryonalcarcinomaandadultstemcells.BiochimBiophysActa‐MolCellRes1803:246–260.

RiccioO, JacobshagenM,GoldingB,VutskitsL, JabaudonD,Hornung JP,Dayer aG(2012) Excess of serotonin affects neocortical pyramidal neuron migration.TranslPsychiatry2:e95.

RiveraC,VoipioJ,PayneJa,RuusuvuoriE,LahtinenH,LamsaK,PirvolaU,SaarmaM,KailaK (1999)TheK+/Cl‐ co‐transporterKCC2 rendersGABAhyperpolarizingduringneuronalmaturation.Nature397:251–255.

RoccioM, Schmitter D, KnoblochM, Okawa Y, SageD, LutolfMP (2013) Predicting

References 7

105

stem cell fate changes by differential cell cycle progression patterns.Development140:459–470.

Rodrigues ARA, Oertel D (2005) Hyperpolarization‐Activated Currents RegulateExcitability in Stellate Cells of the Mammalian Ventral Cochlear Nucleus. JNeurophysiol95:76–87.

Rodriguez‐Gomez JA, Levitsky KL, Lopez‐Barneo J (2012) T‐type Ca2+ channels inmouse embryonic stem cells:modulation during cell cycle and contribution toself‐renewal.AJPCellPhysiol302:C494–C504.

Rosenberg SS, Spitzer NC (2011) Calcium signaling in neuronal development. ColdSpringHarbPerspectBiol3:a004259.

Ross ME, Walsh CA (2001) Human brain malformations and their lessons forneuronalmigration.AnnuRevNeurosci24:1041–1070.

Rowitch DH, Kriegstein AR (2010) Developmental genetics of vertebrate glial‐cellspecification.Nature468:214–222.

RuegerMA,BackesH,WalbererM,NeumaierB,UllrichR, SimardML,EmigB,FinkGR,HoehnM,GrafR, SchroeterM (2010)Noninvasive ImagingofEndogenousNeural Stem Cell Mobilization In Vivo Using Positron Emission Tomography. JNeurosci30:6454–6460.

SachsHG,StambrookPJ,EbertJD(1974)Changesinmembranepotentialduringthecellcycle.ExpCellRes83:362–366.

Sakaue‐SawanoA,KurokawaH,MorimuraT,HanyuA,HamaH,OsawaH,KashiwagiS, FukamiK,Miyata T,MiyoshiH, Imamura T,OgawaM,MasaiH,MiyawakiA(2008) Visualizing Spatiotemporal Dynamics of Multicellular Cell‐CycleProgression.Cell132:487–498.

Sánchez‐Alonso JL, Halliwell JV, Colino A (2008) ZD 7288 inhibits T‐type calciumcurrentinrathippocampalpyramidalcells.NeurosciLett439:275–280.

SandkeS(2006)UntersuchungenneuronalerH‐KanalAktivitätmitHilfe transgenerMäuse(Musmusculus).Dissertation,UniversitätHamburg.

Santoro B, Chen S, Luthi a, Pavlidis P, Shumyatsky GP, Tibbs GR, Siegelbaum S a(2000) Molecular and functional heterogeneity of hyperpolarization‐activatedpacemakerchannelsinthemouseCNS.JNeurosci20:5264–5275.

SantoroB,Grant SG,BartschD,KandelER (1997) Interactive cloningwith the SH3domain of N‐src identifies a new brain specific ion channel protein, withhomologytoeagandcyclicnucleotide‐gatedchannels.ProcNatlAcadSciUSA94:14815–14820.

SantoroB,HuL,LiuH,SaponaroA,PianP,PiskorowskiRA,MoroniA,SiegelbaumSA(2011) TRIP8b Regulates HCN1 Channel Trafficking and Gating through Two

7 References

106

DistinctC‐TerminalInteractionSites.JNeurosci31:4074–4086.

SantoroB,LeeJY,EnglotDJ,GildersleeveS,PiskorowskiRA,SiegelbaumSA,WinawerMR,BlumenfeldH(2010)Increasedseizureseverityandseizure‐relateddeathinmicelackingHCN1channels.Epilepsia51:1624–1627.

Santoro B, Liu DT, Yao H, Bartsch D, Kandel ER, Siegelbaum SA, Tibbs GR (1998)Identification of a gene encoding a hyperpolarization‐activated pacemakerchannelofbrain.Cell93:717–729.

SantoroB,PiskorowskiRA,PianP,HuL,LiuH,SiegelbaumSA(2009)TRIP8bSpliceVariants Form a Family of Auxiliary Subunits that Regulate Gating andTraffickingofHCNChannelsintheBrain.Neuron62:802–813.

Santoro B, Wainger BJ, Siegelbaum SA (2004) Regulation of HCN Channel SurfaceExpression by a Novel C‐Terminal Protein‐Protein Interaction. J Neurosci24:10750–10762.

Scarpulla RC (2008) Nuclear Control of Respiratory Chain Expression by NuclearRespiratoryFactorsandPGC‐1‐RelatedCoactivator.AnnNYAcadSci1147:321–334.

Schaarschmidt G, Wegner F, Schwarz SC, Schmidt H, Schwarz J (2009)Characterization of Voltage‐Gated Potassium Channels in Human NeuralProgenitorCellsLindenR,ed.PLoSOne4:e6168.

SchluscheAK(2011)MorphologischeCharakterisierungzweierH‐StromdefizienterMauslinien.Diplomarbeit,Philipps‐UniversitätMarburg.

SchmalbachB,LepsveridzeE,DjogoN,PapashviliG,KuangF,Leshchyns’kaI,SytnykV,NikonenkoAG,DityatevA,JakovcevskiI,SchachnerM(2015)Age‐dependentloss of parvalbumin‐expressing hippocampal interneurons inmice deficient inCHL1,amentalretardationandschizophreniasusceptibilitygene. JNeurochem135:830–844.

Schulze‐Bahr E, Neu A, Friederich P, Kaupp UB, Breithardt G, Pongs O, Isbrandt D(2003) Pacemaker channel dysfunction in a patientwith sinus node disease. JClinInvest111:1537–1545.

SeifertR, Scholten a, GaussR,Mincheva a, Lichter P,KauppUB (1999)Molecularcharacterization of a slowly gating human hyperpolarization‐activated channelpredominantlyexpressedinthalamus,heart,andtestis.ProcNatlAcadSciUSA96:9391–9396.

SeoH,SeolM‐J,LeeK (2015)Differentialexpressionofhyperpolarization‐activatedcyclicnucleotide‐gatedchannelsubunitsduringhippocampaldevelopmentinthemouse.MolBrain8:13.

Shitamukai A, Konno D, Matsuzaki F (2011) Oblique Radial Glial Divisions in theDeveloping Mouse Neocortex Induce Self‐Renewing Progenitors outside the

References 7

107

GerminalZoneThatResemblePrimateOuterSubventricularZoneProgenitors.JNeurosci31:3683–3695.

SiddiqiF,ChenF,Arona.W,FiondellaCG,PatelK,LoTurcoJJ(2014)FateMappingbyPiggyBac Transposase Reveals That Neocortical GLAST+ Progenitors GenerateMore Astrocytes Than Nestin+ Progenitors in Rat Neocortex. Cereb Cortex24:508–520.

SpitzerNC(2006)Electricalactivityinearlyneuronaldevelopment.Nature444:707–712.

StevensDR,SeifertR,BufeB,MüllerF,KremmerE,GaussR,MeyerhofW,KauppUB,Lindemann B (2001) Hyperpolarization‐activated channels HCN1 and HCN4mediateresponsestosourstimuli.Nature413:631–635.

StieberJ,HerrmannS,FeilS,LösterJ,FeilR,BielM,HofmannF,LudwigA(2003)Thehyperpolarization‐activated channel HCN4 is required for the generation ofpacemaker action potentials in the embryonic heart. Proc Natl Acad Sci U S A100:15235–15240.

StieberJ,StöcklG,HerrmannS,HassfurthB,HofmannF(2005)FunctionalexpressionofthehumanHCN3channel.JBiolChem280:34635–34643.

StonehouseAH,GrubbBD,PringleJH,NormanRI,StanfieldPR,BrammarWJ(2003)Nuclear Immunostaining in Rat Neuronal Cells Using Two Anti‐Kir2.2 IonChannelPolyclonalAntibodies.JMolNeurosci20:189–194.

Strick R, Strissel PL, Gavrilov K, Levi‐Setti R (2001) Cation–chromatin binding asshown by ion microscopy is essential for the structural integrity ofchromosomes.JCellBiol155:899–910.

SundelacruzS,LevinM,KaplanDL(2008)MembranePotentialControlsAdipogenicand Osteogenic Differentiation of Mesenchymal Stem Cells Giannobile W, ed.PLoSOne3:e3737.

Sundelacruz S, Levin M, Kaplan DL (2013) Depolarization Alters Phenotype,Maintains Plasticity of Predifferentiated Mesenchymal Stem Cells. Tissue EngPartA19:1889–1908.

SurgesR,BrewsterAL,BenderRA,BeckH,FeuersteinTJ,BaramTZ(2006)Regulatedexpression of HCN channels and cAMP levels shape the properties of the hcurrentindevelopingrathippocampus.EurJNeurosci24:94–104.

Surges R, Freiman TM, Feuerstein TJ (2003) Gabapentin increases thehyperpolarization‐activated cation current Ih in rat CA1 pyramidal cells.Epilepsia44:150–156.

TakahashiT,NowakowskiRS,CavinessVS(1993)Cellcycleparametersandpatternsof nuclearmovement in theneocortical proliferative zoneof the fetalmouse. JNeurosci13:820–833.

7 References

108

Takahashi T, Nowakowski RS, Caviness VS (1995) The cell cycle of thepseudostratifiedventricularepitheliumoftheembryonicmurinecerebralwall.JNeurosci15:6046–6057.

Tamm C, Pijuan Galitó S, Annerén C (2013) A Comparative Study of Protocols forMouseEmbryonicStemCellCulturingMilstoneDS,ed.PLoSOne8:e81156.

TangB,SanderT,CravenKB,HempelmannA,EscaygA(2008)Mutationanalysisofthe hyperpolarization‐activated cyclic nucleotide‐gated channels HCN1 andHCN2inidiopathicgeneralizedepilepsy.NeurobiolDis29:59–70.

Tang F, Dent EW, Kalil K (2003) Spontaneous calcium transients in developingcorticalneuronsregulateaxonoutgrowth.JNeurosci23:927–936.

TaniguchiY,Young‐PearseT,SawaA,KamiyaA(2012)Inuteroelectroporationasatoolforgeneticmanipulationinvivotostudypsychiatricdisorders:fromgenestocircuitsandbehaviors.Neuroscientist18:169–179.

TavernaE,GötzM,HuttnerWB(2014)TheCellBiologyofNeurogenesis:TowardanUnderstandingoftheDevelopmentandEvolutionoftheNeocortex.

Toth AB, Shum AK, Prakriya M (2016) Regulation of neurogenesis by calciumsignaling.CellCalcium59:124–134.

TylerWA,HaydarTF(2013)Multiplexgeneticfatemappingrevealsanovelrouteofneocorticalneurogenesis,whichisalteredintheTs65Dnmousemodelofdownsyndrome.AnnInternMed158:5106–5119.

UlensC,TytgatJ(2001)FunctionalHeteromerizationofHCN1andHCN2PacemakerChannels.JBiolChem276:6069–6072.

UrregoD,TomczakAP,ZahedF,StuhmerW,PardoLA(2014)Potassiumchannelsincell cycle and cell proliferation. Philos Trans R Soc B Biol Sci 369:20130094–20130094.

Vallejo M (2009) PACAP signaling to DREAM: A cAMP‐dependent pathway thatregulatescorticalastrogliogenesis.MolNeurobiol39:90–100.

Vasilyev D V, Barish ME (2002) Postnatal development of the hyperpolarization‐activated excitatory current Ih in mouse hippocampal pyramidal neurons. JNeurosci22:8992–9004.

VerkerkAO,WildersR (2015)Pacemakeractivityof thehumansinoatrialnode:Anupdate on the effects ofmutations in hcn4 on the hyperpolarization‐activatedcurrent.IntJMolSci16:3071–3094.

Wahl‐Schott C, Baumann L, Zong X, BielM (2005) An arginine residue in the poreregion is a key determinant of chloride dependence in cardiac pacemakerchannels.JBiolChem280:13694–13700.

Wahl‐Schott C, Biel M (2009) HCN channels: Structure, cellular regulation and

References 7

109

physiologicalfunction.CellMolLifeSci66:470–494.

Wainger BJ, DeGennaro M, Santoro B, Siegelbaum S a, Tibbs GR (2001) MolecularmechanismofcAMPmodulationofHCNpacemakerchannels.Nature411:805–810.

WangDD,KriegsteinAR(2009)GABARegulatesExcitatorySynapseFormationintheNeocortexviaNMDAReceptorActivation.JNeurosci28:5547–5558.

WangK,XueT,TsangS‐Y,VanHuizenR,WongCW,LaiKW,YeZ,ChengL,AuKW,ZhangJ,LiG‐R,LauC‐P,TseH‐F,LiRa(2005)Electrophysiologicalpropertiesofpluripotenthumanandmouseembryonicstemcells.StemCells23:1526–1534.

WangX,TsaiJ‐W,LaMonicaB,KriegsteinAR(2011)Anewsubtypeofprogenitorcellinthemouseembryonicneocortex.NatNeurosci14:555–561.

Weissman T a, Riquelme P a, Ivic L, Flint AC, Kriegstein AR (2004) Calciumwavespropagatethroughradialglialcellsandmodulateproliferationinthedevelopingneocortex.Neuron43:647–661.

Wicks NL, Wong T, Sun J, Madden Z, Young EC (2010) Cytoplasmic cAMP‐sensingdomain of hyperpolarization‐activated cation (HCN) channels uses twostructurallydistinctmechanismstoregulatevoltagegating.ProcNatlAcadSciUSA108:609–614.

Wierschke S, Lehmann TN, Dehnicke C, Horn P, Nitsch R, Deisz RA (2010)Hyperpolarization‐activated cation currents in human epileptogenic neocortex.Epilepsia51:404–414.

WilliamsSR,StuartGJ(2000)SiteindependenceofEPSPtimecourseismediatedbydendriticI(h)inneocorticalpyramidalneurons.JNeurophysiol83:3177–3182.

WonderlinWF,WoodforkKA,StroblJS(1995)ChangesinmembranepotentialduringtheprogressionofMCF‐7humanmammarytumorcellsthroughthecellcycle.JCellPhysiol165:177–185.

Wu X, Liao L, Liu X, Luo F, Yang T, Li C (2012) Is ZD7288 a selective blocker ofhyperpolarization‐activated cyclic nucleotide‐gated channel currents? Channels(Austin)6:438–442.

XueT,MarbánE,LiRA(2002)Dominant‐negativesuppressionofHCN1‐andHCN2‐encoded pacemaker currents by an engineered HCN1 construct: insights intostructure‐functionrelationshipsandmultimerization.CircRes90:1267–1273.

YangM,BrackenburyWJ(2013)Membranepotentialandcancerprogression.FrontPhysiol4:185.

Yang SJ, Liang HL, Ning G, Wong‐Riley MTT (2004) Ultrastructural study ofdepolarization‐induced translocation of NRF‐2 transcription factor in culturedratvisualcorticalneurons.EurJNeurosci19:1153–1162.

7 References

110

Yasuda T, Bartlett PF, Adams DJ (2008) Kir and Kv channels regulate electricalproperties and proliferation of adult neural precursor cells. Mol Cell Neurosci37:284–297.

YingS‐W,TibbsGR,PicolloA,AbbasSY,SanfordRL,AccardiA,HofmannF,LudwigA,Goldstein P a (2011) PIP2‐Mediated HCN3 Channel Gating Is Crucial forRhythmic Burst Firing in Thalamic Intergeniculate Leaflet Neurons. J Neurosci31:10412–10423.

Ying SW, Kanda VA, Hu Z, Purtell K, King EC, Abbott GW, Goldstein PA (2012)Targeted deletion of Kcne2 impairs HCN channel function in mousethalamocorticalcircuits.PLoSOne7.

Yoshinaka‐Niitsu A, Yamagaki T, Harada M, Tachibana K (2012) Solution NMRanalysis of the binding mechanism of DIVS6 model peptides of voltage‐gatedsodium channels and the lipid soluble alkaloid veratridine. Bioorg Med Chem20:2796–2802.

ZagottaWN, Olivier NB, Black KD, Young EC, Olson R, Gouaux E (2003) StructuralbasisformodulationandagonistspecificityofHCNpacemakerchannels.Nature425:200–205.

ZhouY,WongC‐O,ChoK‐j.,vanderHoevenD,LiangH,ThakurDP,LuoJ,BabicM,ZinsmaierKE, ZhuMX,HuH, VenkatachalamK,Hancock JF (2015)Membranepotential modulates plasma membrane phospholipid dynamics and K‐Rassignaling.Science(80‐)349:873–876.

ZollesG,KlöckerN,WenzelD,Weisser‐ThomasJ,FleischmannBK,RoeperJ,FaklerB(2006) Pacemaking by HCN Channels Requires Interaction withPhosphoinositides.Neuron52:1027–1036.

ZollesG,WenzelD,BildlW,SchulteU,HofmannA,MüllerCS,ThumfartJO,VlachosA,Deller T, Pfeifer A, Fleischmann BK, Roeper J, Fakler B, Klöcker N (2009)Associationwith theAuxiliarySubunitPEX5R/Trip8bControlsResponsivenessofHCNChannelstocAMPandAdrenergicStimulation.Neuron62:814–825.

ZongX,EckertC,YuanH,Wahl‐SchottC,AbichtH,FangL,LiR,MistrikP,GerstnerA,Much B, Baumann L, Michalakis S, Zeng R, Chen Z, Biel M (2005) A novelmechanismofmodulationofhyperpolarization‐activatedcyclicnucleotide‐gatedchannelsbySrckinase.JBiolChem280:34224–34232.

Zong X, Stieber J, Ludwig A, Hofmann F, Biel M (2001) A Single Histidine ResidueDetermines the pH Sensitivity of the Pacemaker Channel HCN2. J Biol Chem276:6313–6319.

Listofabbreviations 8

111

8 Listofabbreviations

3D threedimensional

aIP apicalintermediateprogenitors

AMPA α‐amino‐3‐hydroxy‐5‐methyl‐4‐isoxazolepropionicacid

aRG apicalRGs

BAC bacterialartificialchromosome

BCA bicinchoninicacidassay

bIP basalintermediateprogenitors

BrdU 5‐Bromo‐2’‐deoxyuridine

bRG basalRGs

BSA bovineserumalbumin

CA cornusammonisregion

CAG cytomegalovirusenhancercombinedwithchickenbetaactin

CaSR calcium‐sensingreceptor

CHO chinesehamsterovary

CISS constructiveinterferenceinsteadystate

CNBD cyclicnucleotide‐bindingdomain

CNPAse 2',3'‐cyclic‐nucleotide3'‐phosphodiesterase

con controlmice

CP corticalplate

CREB cAMPresponseelement‐bindingprotein

CT cyclethreshold

dox doxycyclineuntilP0

E embryonalday


112

EAG ether‐à‐go‐gopotassiumchannel

ER endoplasmaticreticulum

ERK extracellularsignal‐regulatedkinase

FACS fluorescence‐activatedcellsorting

FGF fibroblastgrowthfactor

FOV fieldofview

FPKM fragmentsperkilobasepermillionreads

FUCCI fluorescenceubiquitinationcellcycleindicator

GABA γ‐aminobutyricacid

GC guanine‐cytosine

GE ganglioniceminence

GFAP glialfibrillaryacidicprotein

GFP greenfluorescentprotein

gIP glialprogenitorcells

GYG glycine–tyrosine–glycinemotif

h human

HA Hemagglutinin

HCN hyperpolarization‐activated cyclic nucleotide‐binding non‐selectivecation

HCN‐DN dominantnegativeHCNsubunit

HCN‐WT wildtypeHCN

HRP horseradishperoxidase

i.p. intraperitoneal

IA transient,inactivatingpotassiumcurrent

IC50 halfmaximalinhibitoryconcentration

IDR delayedrectifyingpotassiumcurrent

Listofabbreviations 8

113

IgG immunoglobulinG

Ih hyperpolarizationactivatedcurrent(HCNmediated)

Iins instantaneouscurrent;voltage‐insensitive

IIR inwardlyrectifyingpotassiumcurrent

IP3 inositol1,4,5‐trisphosphate

IRES internalribosomalentrysite

Iss steady‐statecurrent

IUE inuteroelectroporation

IZ intermediatezone

KCC2 potassium/chlorideco‐transporter2

KCR1 potassiumchannelregulatorprotein1

KCTD3 potassiumchanneltetramerization‐domaincontainingprotein3

KID1 kinaseinducedupondepolarization1

LDH lactatedehydrogenase

MAPK mitogen‐activatedproteinkinase

MGE medialganglioniceminence

MiRP1 MinK‐relatedprotein1

mut mutantmice

MZ marginalzone

NE neuroepithelium

NeuN neuronalnucleusantigen

NGS nextgenerationsequencing

NMDA N‐methyl‐D‐aspartate

Nrf2 thenuclearrespiratoryfactor2

NSC neuralstemcellculture


114

OCT optimalcuttingtemperaturecompound

OD opticaldensity

P postnatalday

PACAP pituitaryadenylatecyclase‐activatingpolypeptide

PB phosphatebuffer

PBS phosphatebufferedsaline

PCR polymerasechainreactions

PFA paraformaldehyde

PI propidiumiodid

PIP2 phosphatidylinositol4,5‐bisphosphate

Ptet tetracyclinetransactivatorresponsiveelement

RG radialglialcells

ROI regionofinterest

S transmembranedomains

SVZ subventricularzone

TBS Trisbufferedsaline

tdTomato tandemdimerTomato

TE echotime

TR repetitiontime

TRIP8b tetratricopeptiderepeatcontainingRab8binteractingprotein

tTA tetracyclinetransactivator

TuJ1 classIIIβ‐tubulin

VGCC voltage‐gatedcalciumchannels

VZ ventricularzone

wpc weekspostconception

Acknowledgement 9

115

9 Acknowledgement

Iwould like toparticularly thankProfessorU.B.Kaupp for reviewing thisdoctoralwork and support me with stimulating discussions and remarks. Professor M.Paulsson belongs my heartily gratitude for his unconventionally support andreviewingmydoctoralthesis.

Furthermore, I would like to express my gratitude to Professor D. Isbrandt, mysupervisor, for giving me the opportunity to work in his lab at this fascinatingresearch project, and his scientific and personal support during the years andespeciallyduringthemovementfromHamburgtoCologne.

MysincerethanksgoestoProfessorS.Schoch‐McGovernforhersupportasmemberof my thesis committee, the generous supply with constructs for further in uteroelectroporationexperiments,andheradviceandendorsementinmyfuturescientificsteps.

IhadthegreatlucktobepartofafruitfulcollaborationwithDrS.VayandPD.Dr.A.RügerfromtheresearchgroupNeuralStemCellsoftheUniversityHospitalCologne.Dr.Vaysexpertiseincorticalstemcellcultureandrelatedexperimentshelpedaddingvaluablecontributiontothiswork.

Dr.M.FloriofromtheHuttnerlaboratoryattheMax‐PlankInstituteinDresdenkindlyprovidedmewithNGSdata.

IwasgreatlyadvisedandsupportedattheMRIanalysisbyDr.JSedlacik.

Iamgratefulofbeingpartof theENPTeamandIwould like to thankeveryoneofthem for their scientific and personal support during the years. I would like toespeciallythankDr.I.Jakovcevskiforintroducingmetotheembryonicworldofthebrainandsupervisingmeinthelastyearsofthethesis.ToDr.W.Fazelibelongsmygratitude forproof reading thisdoctoralworkandall thegreatpep talks. Iheartilythank K. Sauter for sharing her tips and tricks in the lab with me, her endlessknowledgeofdiversemethodsandmakingincubationstimessomuchfun.

LastbutnotleastIwouldliketothankmyfamilyandfriendsforwatchingmybackandkeepingmychinup.

10 Erklärung

116

10 Erklärung

Ich versichere, dass ichdie vonmir vorgelegteDissertation selbständig angefertigt,die benutzten Quellen und Hilfsmittel vollständig angegeben und die Stellen derArbeit−einschließlichTabellen,KartenundAbbildungen−,dieanderenWerkenimWortlaut oder dem Sinn nach entnommen sind, in jedem Einzelfall als Entlehnungkenntlichgemachthabe;dassdieseDissertationnochkeineranderenFakultätoderUniversitätzurPrüfungvorgelegenhat;dasssie−abgesehenvonuntenangegebenenTeilpublikationen−nochnichtveroffentlichtwordenist,sowie,dassicheinesolcheVeröffentlichung vor Abschluss des Promotionsverfahrens nicht vornehmenwerde.DieBestimmungenderPromotionsordnungsindmirbekannt.DievonmirvorgelegteDissertationistvonHerrnProfessorDr.U.B.Kauppbetreutworden.

The role of HCN channels in the prenatal development of ...

Documents

Transcript of The role of HCN channels in the prenatal development of ...