Sandwich cortical lamination and single-cell analysis ... · 8/20/2019 · 1 Title: Sandwich...

1

Title:Sandwichcortical laminationandsingle-cellanalysisdecodesthedevelopingspatialprocessingsystem.Authors:Yong Liu1,9, Tobias Bergmann1,9, Julie Lee3, Ulrich Pfisterer4, Louis-FrancoisHandfield5, Yuki Mori6, Andrea Asenjo-Martinez4, Irene Lisa-Vargas4, Stefan ESeemann2,JimmyTszHangLee5,NikolaosPatikis5,JuanMiguelPeralvoVidal1,MariaPihl1,BirgitteRahbekKornum3,7,PrebenDybdahlThomsen1,PoulHyttel1,MennoPWitter8, Konstantin Khodosevich4, Jan Gorodkin2, Martin Hemberg5, Tune H Pers3andVanessaJaneHall*1Affiliations:1Department of Veterinary and Animal Sciences, Faculty of Health and MedicalSciences; University of Copenhagen, Gronnegaardsvej 7, Frederiksberg C, 1870,Denmark.2Centerfornon-codingRNAinTechnologyandHealth,UniversityofCopenhagenGroennegaardsvej3,1870FrederiksbergC,Denmark3Novo Nordisk Foundation Center for BasicMetabolic Research, Faculty of HealthandMedicalSciences;UniversityofCopenhagen,CopenhagenN,2200,Denmark.4Biotech Research and Innovation Centre (BRIC), Faculty of Health and MedicalSciences;UniversityofCopenhagen,CopenhagenN,2200,Denmark5WellcomeSangerInstitute,WellcomeGenomeCampus,HinxtonCB101SA,UnitedKingdom.6Center for Translational Neuromedicine, Faculty of Health andMedical Sciences;UniversityofCopenhagen,CopenhagenN,2200,Denmark7DepartmentofNeuroscience,FacultyofHealthandMedicalSciences;UniversityofCopenhagen,CopenhagenN,2200,Denmark.8Kavli InstituteforSystemsNeuroscience,FacultyofMedicineandHealthSciences;NorwegianUniversityofScienceandTechnology,Trondheim,7030,Norway.AuthorListFootnotes:9Theseauthorscontributedequally*CorrespondingAuthorandleadcontact:VanessaJaneHallContactinfo:VanessaJaneHall:vh@sund.ku.dkSummaryTheentorhinalcortexconsistsofseveralimportantcelltypesincluding,thegridcells,speed cells, border cells and head-direction cells and is important for memory,spatialnavigationandperceptionoftime.Here,wetraceindetailthedevelopmentoftheentorhinalcortex.Usingsingle-cellprofilingweprovideuniquetranscriptionalsignaturesforglia,excitatoryandinhibitoryneuronsexistingintheregion,includingRELN+cellsinlayer(L)IIandsuperficialpyramidalneurons.Weidentifiedasandwichlayeredcortex,whereLIIemergespriortoLIIIandsuperficialcellsmaintainadeep

not certified by peer review) is the author/funder. All rights reserved. No reuse allowed without permission. The copyright holder for this preprint (which wasthis version posted August 20, 2019. ; https://doi.org/10.1101/738443doi: bioRxiv preprint

https://doi.org/10.1101/738443

2

layermolecularidentityafterbirth.Ourfindingscontributetotheunderstandingofthe formation of the brain’s cognitivememory and spatial processing system andprovides insight into the transcriptional identity and spatial position of theentorhinalcells.KeywordsEntorhinalcortex,development,stellatecell,speedcell,single-cellRNAsequencing,corticallamination,neurogenesis,gliogenesis,ventraltelencephalonIntroductionUncoveringthetemporaleventsintheemergenceofthecelltypesinthedevelopingentorhinal cortex (EC) is paramount for understanding the formation of cognitivememory and spatial processing. The EChas a pertinent role in thesememory andnavigationalprocesses,which likelydependson its intrinsicorganizationaswellasextrinsicconnectivity.Corticalprojectionstargetamongstothers,thesubiculumandadjacent parahippocampal regions, ventral retrosplenial cortex, medial prefrontalcortex,contralateralcortexandolfactoryareas.Subcortically,theECconnectstotheseptum,thediagonalbandofBroca,nucleusaccumbens,claustrum,theamygdaloidcomplex, as well as midline thalamic domains (Witter and Groenewegen, 1986).Specificnavigation-relatedfunctionshavebeendiscoveredintheECincludingobjectrecognition (Ridley et al., 1988), processing speedmovement (Kropff et al., 2015),processing location (Quirk et al., 1992), processing head-direction (Taube et al.,1990), recognizingproximal borders (Solstadet al., 2008) andprocessingworkingmemory (Olton et al., 1982). These functions signify the importance of the EC inspatial navigation and arise from subsets of firing cells from either themedial EC(MEC) or lateral EC (LEC). Although extensive anatomical and neuronalelectrophysiologicalclassificationhasbeenperformed(CantoandWitter,2012a,b;Canto et al., 2008), the area lacks detailed molecular characterization. Currently,none,oronlyveryfewgenescanbeusedtoidentifythemajorcelltypeswithintheECandthelocationsofthesecellstypesareonlypartiallyunderstood.Forexample,within the MEC, the grid cells are located in both the dorsal and ventral MEC(Stensola andMoser, 2016) and found across all cortical layers, though showing apreference in layer II (Sargolini et al., 2006). The grid cells are considered torepresent principal neurons including both stellate cells and pyramidal neurons(Domnisoruetal.,2013;Rowlandetal.,2018;Schmidt-HieberandHausser,2013).Gene markers that distinguish the stellate cells from the pyramidal neurons areREELIN(RELN)andCALBINDIN(CALB1),respectively(Fuchsetal.,2016;Leitneretal.,2016;Perez-Garciaetal.,2001;Seressetal.,1994;Vargaetal.,2010;Witteretal.,2017).Also,RELNisexpressedinFancellswhichresideinthesuperficiallayeroftheLECandwhichareimportantforobjectrecognition(Germrothetal.,1989;Nilssenetal.,2018).Amorespecializedgridcellalsoexists;theconjunctivegridcellwhichfireswhenratscrossagridvertexuponheadinginaspecificdirection. It isabsentfromLII,mostlypresentinLIIIandLIV,butcanalsobefoundinLVI(Sargolinietal.,2006).Thehead-directioncellsareprincipalneuronsfoundinLIItoLVIandareabsentfromLII(Giocomoetal.,2014;Sargolinietal.,2006).Thebordercellsarealsoconsidered


https://doi.org/10.1101/738443

3

tobeprincipalneuronsandarefoundwithinalltheMEClayers(Solstadetal.,2008)whichoftenoverlapwithastellatecell identity(Tangetal.,2014).Thespeedcellsmap toGABAergic inhibitoryneuronswhichexpressparvalbumin (PVALB), butnotsomatostatin (SST) (Ye et al., 2018). A clearer understanding of the molecularlandscapeintheECmayhelptoreconciletheanatomical,functionalandconnectiveprofiles of the cell types that exist in the EC to allow a better understanding ofspatialnavigationinthebrain.Itisnotonlythespatialdistributionofcelltypesthatremainpoorlyunderstood,butalsothetemporalmaturationoftheECcells.Head-direction isthefirst functiontoemerge,withhead-directioncells firingconsistently in theratalreadybypostnatalday(P)11(Bjerknesetal.,2015).Therecognitionofproximalbordersdevelopslater,withborder cells firingwith adult-likeproperties at P16-18 (Bjerkneset al., 2014).Spatialnavigationariseslast,withgridcellsslowlyformingadult-likefiringpropertiesafter approximately 4 weeks of age (Bjerknes et al., 2014; Langston et al., 2010).Maturationofcell typeswithincortical layersalsoshowstemporaldivergence.Forexample,inLII,CALB1positivepyramidalneuronsmatureearlierthanRELNpositivestellatecells(RayandBrecht,2016).Thiscontrastswiththeknownbirthingtimeofthese cell types in mice, where stellate cells are born two days earlier at E11.Investigating neurogenesis and the emergence of the varying cell types in thedevelopingECcouldhelptouncoverwhetherthebirthingtimeofcelltypesintheEClinkscloselytofunctionalmaturity.ThiswasalsorecentlyproposedbyDonatoandcolleagues(Donatoetal.,2017).In light of difficulties in obtaining human fetal tissues from the second and thirdtrimesterofdevelopment,wefocusedourattentiononstudyingECdevelopmentinanalternatelargemammal,thepig.Thepigdevelopsagyrencephalicbrainsimilartohumans,andfetaltissueiseasilyacquiredfromproductionfarms. Bothannotatedand unannotated developing and adult pig brain anatomical atlases are readilyavailable frommany open sources. A number of publishedMRI atlases also exist(Conrad et al., 2012; Conrad et al., 2014; Saikali et al., 2010;Winter et al., 2011).These resources demonstrate that porcine and human brain anatomies are highlysimilar. Like the human brain, there is little proliferation of neurons in the largedomesticpigbrainafterbirth(Jelsingetal.,2006).Asignificantincreaseinpigbrainsize following birth is strikingly similar to humans (Conrad et al., 2012). Piggestationaldevelopmentisrelativelylongat114dayswhichismorecomparabletohuman gestation length (294 days) compared to 20 days in mice. Given thedivergenceincortexfunctionsbetweenrodentsandhumans(Farretal.,1988)andvariations in the connectivity of the sensory cortical networks betweenmice andhumans in the hippocampal region (Bergmann et al., 2016), the pig constitutes aparticularlymodeltostudyECdevelopment.Here,wepresentathemostcompleteprofiletodateofthedevelopingECfromananatomical, molecular and single-cell level. We have uncovered a unique corticallamination patterning in the EC which differs from the remaining cortex. Weidentified several unique interneuron and excitatory neuron populations andoutlined themolecularphenotypes for all cell typeswithin theMEC, including the


https://doi.org/10.1101/738443

4

speed cells, stellate cells andpyramidal neuronpopulations.Ourdatapresents anunparalleleddescriptionofnoveleventsduringneurogenesisaswellasthetimingintheemergenceandthemolecularprofileofentorhinalcellstosubstantiallyadvancethefieldinthecharacterizationofthedevelopingspatialnavigationsystem.ResultsSubheading 1. Entorhinal cortex development exhibits rapid growth during thesecondtrimesterofgestationGestational age differences and trimester lengths were first compared betweenmouse,manandpigtogainafullunderstandingofthegestationaltimepointsandtransitions in the pig.We recorded the transition of trimester 1 to 2 as the timepoint when neural tube formation and gastrulation is completed, which is arecommended feature thathasbeenused inmouseandman (Pattenetal.,2014)andkeydevelopmental featuressuchas fusionof thepalatewhichcorrespondstoembryonicday(E)50inthepig(Sunetal.,2017)andgonadaldifferentiation,whichoccursatE50inthepig(Ponteloetal.,2018).WethereforededucetheendofthefirsttrimesterinthehumanequalsE50inthepig.Thetransitionfromtrimester2to3marksasignificantandcontinualincreaseinhumanfetalweightuponcompletionoforganogenesis.AsimilarandmarkedincreaseinfetalweightoccursatE70inthepig(Kim,2010).WebelievethatE70marksanequalrepresentativeofthetransitionfromtrimester2to3.Acomparativeoverviewoftrimesterlengthsbetweenspecies,showsthatthepigisabettermodelthanthemouseintermsoftrimesterphasesinman(Figure1A).The porcine brain has a gyrencephalic cortex and a prominent paleopallium withlargeolfactorybulbs.Theolfactorytractterminatesinalargepiriformlobe.TheECislocatedinthepiriformlobe,caudaltotheolfactory(piriform)cortexandthecorticalnucleus of amygdala (COA) and hippocampal transition area (HA) (Figure 1B).Wefirst analyzed the anatomical features of the EC at a late gestational stage andpostnatal brain using cresyl violet staining. At E100 (2 weeks prior to birth) weobservedacytoarchitecturalmatureECwithsimilarmorphologyasthepostnatalEC(FigureS1).LayerIVwasdistinctivelyacellular,whichisatypicalfeatureintheadultECknownasthelaminadissecans.Weidentifiedtheparasubiculum(PaS)medialtotheEC.ThePaScouldbe identifiedatE100fromitssimilarcytoarchitecturetotheEC.OneexceptionwasthatthesomasizesoftheLIIandLIIIcellswithinthePaSweremore equal in size, compared to the EC. In the EC, the LII cells have somasapproximately twice as large as the LIII cell somas (Witter et al., 2017).We couldidentifytheperirhinalcortex(PER)locatedproximallyandlateraltotheECthroughthedisappearanceoflargesuperficialcellsoftheLIItypicaloftheECtogetherwiththedisappearanceofthelaminadissecans(Figure1C).AsearlyasE100,theECcanbe divided into a lateral (LEC) and medial (MEC) subdivision. This is based oncytoarchitecturedifferencesdescribed inother species suchas rodents, dogs, catsandhumans(Gatomeetal.,2010;Insaustietal.,1995;Woznickaetal.,2006;Wyss


https://doi.org/10.1101/738443

5

etal.,1983).ThepigMEC,haddeeperlayersstructuredwithanarrowhomogenousand densely packed cellular layer organized with a columnar appearance. In thetransition into the LEC, the deep layers widened with a lower cell density and adisordered cell deposition. The lamina dissecans in the MEC was distinct andacellular,whileintheLEC,thelaminadissecanswasmorediffuse,withmorecellularinfiltration (Figure1CandS1).TheMECgraduallyoccupiedtheentiremediolateralentityalongtherostrocaudalaxisofthepiriformlobe.TheMECcaudallyborderstothe neocortex as the piriform lobe disappears. This boundary was traced byimmunolabelingforparvalbumin(PVALB)positiveinterneurons(FigureS2A),asseenin the rat (Wouterlood et al., 1995). We observed a sulcus forming at the mid-rostrocaudalextentoftheEC,persistingtothecaudalendofthepiriformlobewhichwasalsovisiblemacroscopically(Figure1B).Thissulcus,knownasthesagittalsulcus(S.Sag)hasbeenpreviouslyreportedinthedomestic(HolmandGeneser,1989)andGöttingenminipig(Bjarkametal.,2017),butnotinothergyrencephalicspecies,suchasdogs,cats,andhumans(Insaustietal.,1995;Woznickaetal.,2006;Wyssetal.,1983).WeobservedtheS.SagtocoincidewiththeappearancewiththeMEContherostrocaudal axis (Figure S1), which provided a unique surface landmark formicroscopicallyandmacroscopicallyassessingtheMECandLECborders.Interestingly,weobservedahighnumberofoligodendroglia-likecellspresentinthesuperficial layersof the LECand in theadjacentperirhinal cortex (PER),whichhasnot been previously highlighted in other anatomical descriptions of the EC. Thesecells were proximal and clustered around large entorhinal cells (Figure. 1D). Weobserved enrichment of these oligodendroglia-like cells in the superficial layers oftheLECandPERfromE80throughdevelopment.Thesecellswerestillapparent intheLECandPERatP75(Figure1D).ForearlierstagesthanE80similarglia-likecellswereobservedinalllayersandthereforepotentialoligodendrogliacellscouldnotbedistinguished (Figure S2B). Morphologically, these appeared to be pan-neuronaloligodendrocyte precursor cells (OPCs)with clearNissl free cytoplasm and a largeround Nissl stained nucleus (Garcia-Cabezas et al., 2016) (Figure 1E). In themostlateral part of the LEC, these OPC-like cells were evenly distributed within thesuperficial layer, but converged into distinct islets towards the MEC. The isletsdisappearedat theborderof the LEC-MECand the clusteringof theOPC-like cellswasnotevidentintheMEC.WeimmunolabeledtheECwiththeOPCmarker,OLIG2(FigureS2D),andthesecellsweredetectedfromE60onwards.ExpressionofOLIG2wasalso foundacrossall layers (FigureS2E).Theseoligodendroglial cellswerenotobserved in the neighbouring neocortex regions and have not been reported inother species, indicating that thismaybeaunique featureof theporcineLECandadjacentPER(Figure1D-E).WesubsequentlyanalyzedthetimingoftheemergenceanddevelopmentoftheECusingcresylvioletstaining(E50-P75)andstructuralpostmortemMRI(E60-P75)fromgestationalages to thepostnatalbrain.No featuresofanECcouldbedeterminedpriortoE50.AtE50,weidentifiedprominentandlargesomaentorhinal-likecellsinthe superficial layer of the ventral telencephalon before any distinct layer wasformed(FigureS2C).Thethree-layeredcortexbecamesix-layeredbyE60(FigureS1)and at the same time the lamina dissecans was detected (Figure S1). At E60, we


https://doi.org/10.1101/738443

6

wereabletodelineatetheMECfromtheLECbasedonthedepositionofthedeeplayercellsandthehighercellularityofthelaminadissecansintheLEC.Together,thisindicatesthattheECcorticallayersareformedduringthesecondtrimesterbetweenE50andE60inthepig.AtE70(lateinthesecondtrimester),weobservedtheS.Sagformationmicroscopicallyandcells in thesuperficialLIIwereprominent, largeanddarklystainednucleiwithNissl(FigureS1).FromE80,thelargecellsinLIIappearedmoreprominent(FigureS1)andtheS.Sagwasmacroscopicallyvisible(FigureS1).ThepositionalanatomyandrateofECgrowthwasthenevaluatedusingpostmortemstructural-MRI. Brains from E60 to P75 were scanned on a 9.4T preclinical MRIsystem (BioSpec 94/30 USR, Bruker BioSpin, Ettlingen, Germany). The EC wasannotatedfromE60toP75basedonourhistologicaldescriptionsandmacroscopicfeatures(Figure2A).WecalculatedthevolumeoftheECandtheremainingcortexbasedontheMRIannotationsandfoundtheECgrowthwaslinearfromE60toP75(Figure 2B).When we compared the growth rate of the EC to that of the wholecortexof thebrain,we foundthat theEChadasignificantlyhighergrowthrateatE70 during the late second trimester (Figure 2B), suggesting a local and specificdevelopmentalgrowthperiodatthistimepoint.Wewereunabletodetectwhetherthegrowthwasduetoexpansionofthewhiteorgraymatter,duetothelackofhighenoughresolutionfromtheMRIimages.Insummary,weidentifiedprominentOPCsat theearlystageofECdevelopment,wherethecellswereclusteredaround largeentorhinal cells and a significant growth spurt during the late second trimesterattributabletoeithergreyand/orwhitematterexpansion.TofurthervalidatethepigasasuitablemodelforstudyingthedevelopmentofEC,weanalyzed thenumber andanatomical endpointsofwhitematter fiberspassingthrough or emerging from the MEC and LEC using diffusion tensor imaging (DTI)tractography in the postnatal brain (P75). A high number of white matter tractscouldbedetectedemerging fromboth theMEC (median6031±1319)and the LEC(median 3329±4099) (Figure 2C-D, Supplementary Table 1). Analyses of the tractsarising from the EC area revealed the septohippocampal system as the mainprojection site fromboth the LECandMEC,howeverprojectionsalsoextended tothe subiculum, olfactory bulb, the nucleus of the lateral olfactory tract (LOT),amygdala, corpus collosum, putamen (PUT), nucleus accumbens and corpuscollosum(Figure2E-J),Suppvideo1,SuppTable1).ConnectionsbetweentheMECandLECcouldalsobeobserved(Figure2J,Suppvideo1).Regardingconnectivitytothehippocampus,theLECconnectedtothedistalpartofCA1andtheMECmainlyconnectedtotheproximalpartofCA1,withaminornumberofconnectionstoCA2and CA3 (Supp table 1, Supp. video 1). In addition, the MEC connected to thedentategyrus(DG)attheseptalendofthehippocampusandtotheperirhinalcortex(Suppvideo1),whilsttheLECmostlyconnectedtothepostrhinalcortex(Figure2E).Theobservedconnectivitytotheperi-andpost-rhinalcortexconcurswithpreviousstudies reported in the rat (Burwell and Amaral, 1998; Insausti et al., 1997). Ofinterest,wasasmallnumberoffibersthatemergedfromtheMECanddisappearedintothecorpuscollosum(Figure2F,J,Suppvideo1).Sincethebrainswerescannedas semi-hemispheres to obtainmaximal resolution, it was difficult to tracewherethesetractsended.Strikingly,mostofthetractstracedtootherregionsandalarge


https://doi.org/10.1101/738443

7

numberextendedtotherostroventralbrain,whichappearstobetheanteriorvisualcortex (remains unannotated in the pig) and many LEC fibers ended in a slightlycaudallocationtotheMECfibers(SuppVideo1).Tosum,theconnectivitybetweentheECandhippocampus in thepostnatalpigbrainwas remarkably similar to thatwhichhaspreviouslybeenreportedinotherspecies,includinghumans(Kolenkiewiczetal.,2009;Mufsonetal.,1990;Rowlandetal.,2013;Sunetal.,2014;TotterdellandMeredith, 1997;Wilhite et al., 1986;Witter et al., 2017;Witter et al., 1986)whichhelpstoconsolidate itasanexcellentmodelforhumanECconnectivity.Wehavehowever,identifiedanumberofpotentialnewlocationsofECconnectivitythatrequiresfurtheranatomicalanalysis.Subheading2.CorticallaminationofthedevelopingentorhinalcortexissandwichstructuredThetimingofbirthofpyramidalneuronsandstellatecellsdoesnotcoincidedespitethecell’spositionalproximity.WethereforedecidedtotracethecorticallaminationpatterningoftheECandemergenceofcelltypesindetail.Tothisend,weexaminedthe MEC among developing posterior EC from E23 to P75 (Figure S3A). Sinceneurogenesishasbeenpoorlystudiedinthedevelopingpigbrain,wefirstevaluatedthe timing of neurogenesis. Immunohistochemistry of theMECwas performed atover 8 time points from E23 to P75 of development. Using antibodies directedagainst GFAP, FABP7 (BLBP), SOX2 and PAX6,we assessed the presence of neuralepithelium(NE) (PAX6+/FABP7-)atE23andradialglia (GFAP+/FABP7+)atE26,andwealsomeasuredthesizeof theventricularzone(VZ) (PAX6+) (FigureS3B-D).WeidentifiedthattheVZisalreadywell-establishedbyE23andthatitpeakedinsizeatE50,whereby it thendeclineddramatically in sizeatE60 (FigureS3C,D).We theninvestigated the second germinal zonebyperforminghistochemistry using EOMESandPAX6specificantibodies.OurresultsshowedthepresenceofamoderatelysizedlayercontainingdiffusenumbersofEOMES+duringneurogenesis.Thisdiffuselayer,knownas theoutersubventricular zone (OSVZ), isextensive insize inhumansandnon-human primates, but smaller in ferrets and rats (Fietz et al., 2010;Martinez-Cerdenoetal.,2012;Smartetal.,2002).Assessmentandquantificationofthecelltype proportion and population, respectively showed that EOMES+/PAX6+ andEOMES+/PAX6-SVZshowedequalratioofpopulationatE26(FigureS3E,F).AtE33,two distinct layers could be observed above the VZ, a dense inner subventricularzone(ISVZ)containingequalproportionsofEOMES+/PAX6+andEOMES+/PAX6-cellsand a diffuse overlying zone, the OSVZ, containing EOMES+/PAX6+ andEOMES+/PAX6- cells. We observed that the ratio between EOMES+/PAX6+ andEOMES+/PAX6-populationchangedovertime,withEOMES+/PAX6+becomingmoreabundant inboththe ISVZandOSVZ(FigureS3E,F),whenthedistinctmorphologywasdeclinedatE70.TheOSVZ remainedmoderate in size.ConsideringPAX6 isanimportant transcriptional regulator of neurogenesis, with increasing expressionchangingthebalancefromneuralstemcellrenewaltoneurogenesis(Sansometal.,2009), our findings implying that neurogenesismight occur from E33 until shortlyafterE60.


https://doi.org/10.1101/738443

8

Analysisofcortical laminationwasperformedusing immunohistochemicalanalysesofcanonicalmarkersfordeep(CTIP2/BCL11B)andsuperficiallayers(SATB2),aswellas EOMES, RELN and TBR1 across varying times of gestation. During early corticalformation(E21)wedetectedthepresenceofEOMES+andRELN+cellstowardsthepialsurface.Co-expressionofEOMES+/RELN+cellsweredetected2dayslateratE23and expression of TBR1 emerged in some of these cells at E23 with moreRELN+/TBR1+cellsdetectablebyE33 (FigureS3G,H).Webelieve thesecells tobemigrating andmaturing Cajal–Retzius cells (CR cells), which are important for thecortical laminationprocess(Hevneretal.,2003).WethentracedtheemergenceoflocalneuronsusingBCL11BandSATB2.Asexpected,BCL11Bwasthefirstofthetwomarkers tobeexpressedatE33andwas located in thecorticalplate (FigureS4A).Interestingly, a large proportion of the BCL11B+ cells expressing SATB2 in thecytoplasm could be detected in the superficial layer at E39 (Figure S4A). This co-expressioncouldalsobedetectedinthedorsaltelencephalon,butatanearlierstageatE33 (FigureS4A),suggestinga timelineofneurogenesis indifferentareasof thecortex. By E50 a distinct new population of BCL11B cells had emerged at thesuperficial layeroftheECwhichcouldnotbeobservedinthedorsaltelencephalon(Figure 3A, B). These large prominent nuclei resemble the prominent largeentorhinal cells that were observed histologically (Figure S2). In the dorsaltelencephalon, a SATB2+populationemerged in the superficial layer at E50whichremainedpresentuntilgestation(Figure3A,B)whichwasnotdetectedintheEC.Incontrast,intheEC,anewpopulationofBCL11BpositivecellsemergesatE60whichdoesnot lieoverthesuperficial largenucleiBCL11Bpopulation.This indicatesthattheECdoesnotforminaninsideoutlaminationpattern,butthelatestbornneuronsformLIII insteadofLIIneurons.Thispositioning is retaineduntilafterbirth (Figure3C)andisnotnotedinthedorsaltelencephalon(Figure3B).Asthelaminadissecansdevelops, theSATB2+neuronsoriginally foundat the superficial layerat E33 formtheLV.AsmallproportionoftheLIIBCL11B+cellsbegantoexpressRELNatE60withexpressionbecomingmoreabundantinthesuperficialcellsatE70(Figure3B,C).TheRELN+neuronsresidinginboththeMECandLECcontinuedtoexpressBCL11Buntilafter birth (Figure 3D, E). Collectively, our findings suggested that the superficialneuronsoftheECarisefromthefourthwaveoflaminationandthatthefifthwaveoflaminationformsLIII.WethereforeconcludethatlaminationpatterningintheECdiffersremarkablyfromneighbouringareasofthecortex.Subheading3.Single-cellRNAsequencingreveals32cellpopulationsintheMECSingle-cell RNA sequencing (scRNA-seq) is a powerful approach that has led to adeeperunderstandingofthemolecularidentityofseveralcelltypesinvaryingbrainregions.We used thismethodology to investigate the timing in the emergence ofdifferentcelltypesoftheECandtheirmolecularidentityduringdevelopmentandinthepostnatalbrain.WeperformedscRNA-sequsingthe10XGenomicsmicrofluidicsbasedChromiumplatformonwhole-cellsisolatedfromtheECatE50,E60,E70andP75,andisolatednucleiinthelaterstagesofdevelopmentatE70andP75toensureefficientcaptureoftheneuronsinthetissue(Figure4A).Downstreamclusteringand


https://doi.org/10.1101/738443

9

data analysis was performed in Seurat v2 (Satija et al., 2015) (Figure 4A). Weexcludedredbloodcells(expressinghemoglobinsubunitsHBB,HBE1,HBM,HBZandonaverage,only291genespercell)andvascularcells(PDGFRB+/PECAM1+)fromourdatasetwhich left uswith24,294 cells. Confident inourdata,we thenperformedcellclusteringanddetected32distinctcellclusters(Figure4B).Theclusterscouldbedelineated into 6main cell populations demarcated by canonicalmarkers for IPs,excitatory neurons, inhibitory interneurons (IN),microglia, oligodendroglia (Oligo),andglialcells,whichincludedbothastrocytes(AQP4+/GFAP+/GLAST+)andradialglia(HES1+/SOX2+)(Figure4C).Onecluster,namelycluster30,containedasmallnumberof cells (n=70) which contained twice as many transcripts and genes expressed(Figure4C)whichwebelievedtobedoubletsandwhichfailedtoberemovedbythedata-preprocessing.Thisclusterwasnot included infurtheranalyses.Furthermore,we validated our marker-expression driven annotations by projecting our datasetusingscmap(Kiselevetal.,2018)ontoapubliclyavailableandannotateddatasetofthe human embryonic prefrontal cortex (Zhong et al., 2018) and an unpublisheddataset from the human adult middle temporal gyrus (Bioarchive,https://doi.org/10.1101/384826 ) (Figure 4E). We found a high concordancebetweenourannotationandthehumanbraindatasets,despitetheobviousregionaldifferences,with twoexceptions.Adiscrepancywasobserved in theannotationofclusters2and25.WeinitiallyidentifiedtheseasINs,however,thescmapprojectionannotatedthesetwoclustersasexcitatoryneurons(Figure4C,E).Despitethescmapannotation of cluster 25 as excitatory neurons, their expression profile related tothatof INwithexpressionofGAD1/2andabsence inexpressionof vGLUT1/2. Forcluster 2, a closer analysis of canonical markers suggests an ambiguous identity,mainlyatE60,expressingboththeexcitatorymarkersvGLUT1/2andtheINmarkersGAD1/2 (Figure4B,C). TheUMI counts for this clusterwerenormal,meaning it isunlikely to be a doublet cluster and might suggest a unique and ambiguousprogenitorcelltypeintheEC.OurconfidenceinthedatasetallowedustoprogressinassessingthegliaandneuronpopulationstofurtheranalysethegenotypesoftheECcells.Distinguishingsubtypesofspecificcelltypesbasedontheentirepopulationusing Seurat is difficult. Therefore, subclustering and subsequent analyses wereperformedontheoligodendroglia (Figure5),excitatoryneuronsand IPs (Figure5),and IN clusters (Figure 6) to elucidate new cell types within these discretesubpopulations.Subheading4.Single-cellRNAsequencinganalysisrevealsOPCsemerge intheECatE50.WecharacterizedthetranscriptomeofthepredominantpopulationofOLIG2+OPCsfound in the superficial layers of the LEC (Figure 1D-E, S2D-E). Subclustering andanalysis of the oligodendroglia clusters 3, 12, 24, 27 (Figure 5A insert) revealed 5distinct populations (Figure 5A, B). Cluster 30, which grouped with theoligodendrocytesandexpressedMBP(Figure4),wasomittedfromthesub-analysis,asthehighestenrichedgenesinthisclusterwerefoundtobemitochondrialgenes(Figure S5A) and they are likely to be dying cells. Two clusters (OPC1 and 2)expressedtheoligodendrocyteprecursorlineageandearlyoligodendrocytemarkers


https://doi.org/10.1101/738443

10

PDGFRAandOLIG1 (Figure5C),witha largeproportionofthecellsundergoingcellcycle division (50% and 25% respectively, Figure 5D).Wewere unable to identifyOLIG2 in these twoclusters, likelydue to lowcopynumbersandsubsequentgenedropouteffects. However, thisdatasuggestedthesetwoclusterswereprogenitorcells.ClusterOPC1wascomposedoffetalcellsfromE50toE70whilstclusterOPC0also contained adult cells (Figure 5B). We therefore believe that cluster OPC1emerges earlier during development and corresponds to the first wave ofoligodendrogenesis.WebelievethesetobetwoseparatepopulationsofOPCs,withone remaining present until after birth and the other differentiating into pre-myelinating oligodendrocytes (and changing their transcriptional profile). Tocharacterizewhethertheseclustersresidedinspatiallydifferentniches,wefurtherexamined their unique gene signatures (Figure 5C). Cluster OPC0 expressed thegenes LHFPL3 and MMP16, and cluster OPC1, STMN1. These genes have beenpreviouslyidentifiedinOPCs(Artegianietal.,2017;Huetal.,2011;Linetal.,2009;Magrietal.,2014).Interestingly,wefoundtheOPC0markerCNTN1tobeexpressedinthesuperficiallayersinthedevelopingmouseEC(Figure5E,fromtheAllenBrainAtlas(Leinetal.,2007)).WebelievethattheOPC0populationrepresentstheuniquepopulationofOPCsidentifiedhistologicallyinthesuperficiallayersoftheLEC(Figure1D-E).Interestingly,ourdatarevealedanadditionalpopulationwhichconstitutedoffetalcells frommainlyE60;OLI2 (Figure5A-B).Thesecellsexpressedmorematureoligodendrocytemarkersincluding,MBP,CLDN11,GRP17,NKX2-2,andMAG(Figure5C).OLI2cellsexpressedhighlevelsofFYN,predominantlyintheupperlayersofthedevelopingE18.5mouseEC.Together,thisdatasuggeststhatasmallpopulationofmyelinating oligodendrocytes is apparent already in the fetal porcine EC. Theoligodendrocyte clusters OLI3 and OLI4 consisted almost exclusively of adult cellsandwerehighlysimilar inprofile(Figure5B,C).Wespeculatedthesewerematuremyelinatingoligodendrocytes. Theydiffered inproliferation asnearly allOLI3 cellsexpressedG2M/S-phase cycling geneswhileOLI4 cellsweremostly quiescent andexpressiongenesfromtheG0/1-phaseofthecellcycle(Figure5D).TheOLI3uniquegenemarker,OXR1, was expressed in LI-III of the adultmouse ECwhile the OLI4uniquegenemarker,DLG2,wasfoundinthedeeplayers(Figure5E-F).Insummary,ourdataindicatethepresenceof2OPCpopulationsthatresideindifferentlayersofthe EC, one pre-myelinating oligodendrocyte population and two similar matureoligodendrocyteprofiles.Subheading5.Single-cellRNAsequencingrevealsuniquegenesignaturesforstellatecellsandpyramidalneuronpopulationsTo gain a deeper understanding of the temporal pattern of emerging neuronsweperformedsubclusteringontheIPandexcitatoryneuronpopulationsidentifiedfromthe parent dataset and uncovered unique gene expression signatures. Wesubclusteredandanalyzedatotalof7,672cells,includingtheexcitatoryclusters0,5,10,15,and17andIPclusters19,21,22,and23(Figure6A,insert).Thesubclusteringrevealed 13 distinct populations (Figure 6A). Cells from the gestational stagesdominatedthesubsetof IPsandexcitatoryneurons,withonlyvery fewadultcellsincluded (Figure 6B). This is likely due to the overall fewer cells captured in the


https://doi.org/10.1101/738443

11

postnatalbrainsamples.Amongsttheclusters,onepopulationpertainedatypicalIPidentity.Herein,cells incluster IP4expressedthetypical IPmarkersEOMES,SOX2,and NEUROD4 (Figure 6C) (Pollen et al., 2015) and could be detected across allstagesofdevelopment(Figure5B).Ananalysisofcanonicalgenesfor immatureneuronswithapyramidal-like identityrevealed several potential cell populations. Clusters PYR1, PYR0, PYR2, PYR3 andPYR12 were tangentially located on the t-SNE plot (Figure 6A) and co-expressedNRP1 (Figure 6C), which is reportedly expressed in pyramidal precursor cells anddownregulatedinmaturepyramidalneurons(Sanoetal.,2017).Surprisingly,wedidnotobserveanyneuronpopulationsexpressingCALB1,acommonpyramidalneuronmarker reported in rats, mice and humans within the EC (Beall and Lewis, 1992;Diekmann et al., 1994; Fujimaru and Kosaka, 1996). Further analysis ofimmunohistochemicallabelledECdidrevealexpressionofCALB1intheadultECandonlylimitedexpressionatE100(FigureS2A),withnoexpressiondetectableatearlierstages.Similarly,inthemousebrain,afewCALB1positivecellsarepresentintheECat E18.5, while at P4 CALB1 is widely expressed in the EC (Allen Brain Institute,developmental ISH atlas), suggesting that CALB1 is not expressed in the fetal EC.Clusters PYR0 and PYR1 shared similar transcriptional profiles but had divergentNEUROD1expression.ClustersPYR2andPYR3werealsoremarkablysimilar,buthaddivergent ARPP21 expression. We found cluster PYR12 to be an intriguingpopulation. It had an overlapping profile of both the PYR0, PYR1 and PYR2, PYR3clusters, suggestive of an immature pyramidal neuron, however, it also expressedthemorematurepyramidalneuronmarkerEMX1(Chanetal.,2001)andamoderatelevelofRELN (Figure6C) (amarker for thestellatecell) (Perez-Garciaetal.,2001).Additionally,thisclusteralsohadlowexpressionoftheinterneuronmarkersDLX1/5(Figure 6C) andGAD1/2 (data not shown). Further investigations are required toassesstheidentityofclusterPYR12.Webelievewehavecapturedtheprofileofatleast 2 young pyramidal neuron populations, since it is difficult to separate PYR0fromPYR1and to separatePYR2 fromPYR3.Weare confident in this, since thesepopulationswere captured across all gestational ages (Figure 6B).We additionallycapturedayoungneuronpopulationwithbothanexcitatoryandinhibitoryidentity(PYR12). This population of cells with a unique ambiguous transcriptional identityhasnotbeenreportedpreviouslyandlikelybelongstooneoftheuniquecelltypesfoundintheEC,requiringfurtherinvestigation.An analysis of mature pyramidal neuron markers revealed several potentialpopulations,ofwhichtwooverlappedwithaprecursoridentity.ClustersPYR2,PYR3,andPYR7expressedEMX1andtheexcitatorymarkerneuronmarkervGLUT2(Figure6C). Cluster PYR6 expressed the pyramidal neuron marker TBR1 (Englund et al.,2005)andsharedasimilargenesignatureprofiletothatofclusterPYR7(Figure6C).GiventhetangentialshapesofclustersPYR1,PYR0,andPYR2,PYR3andPYR7,PYR6and the tangential change of immature to mixed immature/mature to maturetranscriptional identity it is tempting to speculate the transitional emergence of asinglepyramidalneuron.However,thegenesignatureprofilesofclustersPYR0,PYR1wereremarkablysimilartoeachother,aswasclustersPYR2,PYR3andclustersPYR6,PYR7 and all were present across several gestational ages. Most interestingly,


https://doi.org/10.1101/738443

12

clustersPYR0,PYR2andPYR6wereonlypresentinthepostnatalbrainandthereforewededucetohavecaptured3pyramidalneuronprofilesofimmature(clustersPYR1,PYR3,PYR7)identityandmature(clustersPYR0,PYR2,andPYR6)identity.Validationof two unique gene markers in the Allen Brain Atlas corroborated the putativelocationofthesecells.ThegeneTNRishighlyexpressedinPYR2andPYR3andwasfound in the superficial layers of the mouse EC. The geneNEUROD6was highlyupregulatedinPYR2,PYR3andPYR6,andPYR7clustersandcanbedetectedinboththesuperficialanddeeperlayers,suggestingthatclustersPYR6andPYR7mayresidebothinthesuperficialanddeeperlayersordeeperlayersalone.Takentogether,webelievethatwehaveidentifiedthreepyramidalneurontypesintheECwithuniquespatiallocation.Inadditiontothepyramidalneurons,theECharborstwotypesofprincipalneuronsexpressing RELN, the stellate and fan cells, residing in LII/III in theMEC and LEC,respectively (Witter et al., 2017). Our subclustering analysis revealed 4 distinctexcitatory neuron populations that expressed RELN. Apart from the knownexpression of RELN and absence of CALB1, few other genes are known to beexpressed in these cells (Fuchs et al., 2016; Kitamura et al., 2014;Winterer et al.,2017).TheclustersRELN5,RELN8,RELN10,RELN11expressedRELNandlackedthepyramidal transcription factor, EMX1 (Figure 6C). Cluster RELN8 and RELN11separated distinctly on the t-SNE plot and had distinct gene signatures to theremaining clusters (Figure 6C). In particular, cluster RELN11 was different fromclusterSTE8,sincenearlyallcellswereineitherG2MorS-phase,whilethemajorityofcellsinclusterRELN8wereinG1-phase(FigureS5B).Bothclusterscouldonlybedetectedduringneurogenesis(E50andE60),thereforewebelievethesetobeyoungRELN positive neurons of different identities. In the E18.5 mouse EC, a uniquemarker for the RELN11 cluster,CXCR4,was expressed across all the layers (Figure6D).TheothertwoclustersSTE5andSTE10wereexpressedthroughoutallagesandpersisted in the postnatal adult brain (Figure 6B). They had quite distincttranscriptionalprofilesandseparatedfarfromeachotheronthet-SNEplot(Figure6A-C).Wecouldnot identify the locationofRELN5neuronsusingthemousebrainatlasandthemarkersthatwereuniquelyfoundinourdataset,thereforewecannotdeducethelocationoftheseneurons.ClusterRELN10,expressedmoderatelevelsofRELN and also expressed the marker gene CPNE4, which is expressed in thesuperficial layersof themouseECandmoreenriched in the LEC than theMEC LII(Figure6B-D).NearlyallcellsintheclusterwerefoundtobeinG1/G0-phaseofcell-cycle(FigureS5B).Further,EPHA6whichwasupregulatedinclusterRELN10couldbeobservedintheLII-LIIIintheadultmousebrain,confirmingthelikelylocationoftheRELN10neurons,tobeinthesuperficiallayer.WebelieveclusterRELN10representseithertheadultstellateand/orfancellgenotype.Interestingly,we identifiedauniquepopulationwhichdidnot fit either stellateorpyramidal cell expression profiles, cluster SPE9. This population was negative forvGLUT and SST expression, but expressed RELN, and GAD1/2. Therefore, weidentifiedauniqueexpressionprofile for thispopulation (Figure6C).ExpressionofSFRP2,oneofthecluster-specificgenes,isfoundnotonlyinthesuperficiallayersofthe EC, but across all the layers in adultmice, and across both theMEC and LEC


https://doi.org/10.1101/738443

13

(Figure6D).Wefoundthatmostcellswereundergoingcell-cycledivision,whiletheotherpartof the clusterwere inG1/G0-phase (FigureS5B). This suggests that theclusterincludedbothprogenitorsandmaturestatesofthecelltype.Together,thesefindings concur with current knowledge on speed cells, known to be GABAergic,absentforSSTexpressionandpresentacrossalllayersoftheEC(Kropffetal.,2015;Yeetal.,2018),althoughmarkerexpressionsuggeststhiscelltypeisequallypresentin the LEC. Since no other population expressed this combination ofmarkers, webelievewehaveidentifiedthisuniquecelltypeintheECandprovideauniquegenesignatureforthiscellincludingSFRP2,COL2A1,COL18A1,SIX3andSIX6.TofurtherverifythattheexcitatoryneuroncelltypesweidentifiedfromthepigECare representativeof EC cells in other species,weprojectedour IP andexcitatoryneuron dataset onto a recently published scRNA-seq dataset of the developingentire brain from P2 and P11 mice which included 156,049 single nucleitranscriptomes(Rosenbergetal.,2018)(FigureS6A-B).Theprojectionshowedthatour pyramidal clusters matched closely with themouse cortical and hippocampalpyramidal neurons identified from the Rosenberg dataset. Our remaining neuronclusters matched less well to the data set. The RELN clusters were moreheterogeneousinnature,withnostrongmatchestoanysingleclusterandtheSPE9clustermatchedclosestwithhippocampalpyramidalneuronsandcerebellargranuleprecursors.WebelievethesequencingdepthoftheRosenbergdatasetmaynotbedetailed enough to identify the EC specific neuron clusters, but it helps to verifyusing a whole brain dataset from a different species that our excitatory neuronpopulations,albeitunique,containpyramidal,glutamatergicneurons.Subheading 6 Three somatostatin positive and three somatostatin negativeinterneuronpopulationsaredetectedintheentorhinalcortexTogainabetterunderstandingofthetemporalpatternofemergingINswithintheEC, we sub-clustered a total of 2,942 cells from the dataset which expressedcanonical IN markers DLX1, GAD1 and GAD2 (clusters 2, 25, 18 and 16). Thesubclusteringidentified7clusters(IN0-IN6,Figure7A).Theproportionofcellsfromdifferent developmental ages within each cluster can be seen in Figure 7B. Weevaluated these clusters for the expression of several well-known canonical INmarkersincludingCALB1,CALB2,DLX5,GAD1,GAD2,GATA3,LHX1,LHX5,LHX6,NPY,RELNSST,TACR1(Figure7C)andHTR3A,PVALB,CCK,NOS1,LHX5,andVIP,althoughtheselattergenescouldnotbeidentifiedinthescRNA-seqdataset,likelyduetolowexpression.Of these lattermarkers,PVALBhasbeenparticularly important for theidentificationofINsintheEC.SeveralreferenceshavereportedPVALB+INsandtheirinhibitorygradientsintheEC(Beedetal.,2013;FujimaruandKosaka,1996;Miaoetal.,2017;Willemsetal.,2018;Wouterloodetal.,1995).ToconfirmthatPVALBwasexpressedatlowlevelsduringearlygestation,weperformedimmunohistochemistryanddetectedmoderatePVALB+ inthesuperficial layersfromE100(Figure7G)andonwards(P75)(FigureS2A).


https://doi.org/10.1101/738443

14

Wethenidentifieduniquetranscriptionalsignaturesforeachcluster(Figure7D).TheIN0-IN2 populations expressed SST, whereas the IN3-IN6 populationswere absentforSST(Figure7C).ImmunolabelingoftheECbrainrevealedhighexpressionofSSTacross all the EC layers at E60 (Figure 7F). IN0 and IN1 had relatively similartranscriptionalprofiles,withsomenotablesmalltranscriptionaldifferences.IN0hadhigher SST expression and lower expression of RELN. IN0 also expressed GAD2(Figure 7C). Both clusters uniquely expressed a gene, SNCB, which was identifiedacross LII to LVI in the adultmousebrain.Another gene,RESP18wasmorehighlyexpressedin IN1andwas localizedtothesuperficial layersanddeeplayers(Figure7E), suggesting IN1 is not found in themiddle layers. However, given the relativesimilaritiesofexpressionbetween IN0and IN1 (Figure7D), it cannotbeconcludedthat IN0and IN1areseparatecellpopulations.Theother INpopulationsappearedmore transcriptionally unique. The IN2 population was captured across all timepoints and had high expression of RELN (Figure 7B,C). Expression of the GABABreceptorsubunitgeneKCTD8ishighlyupregulatedinIN2andisexpressedacrossallthe layers in the adult mouse brain (Figure 7E) suggesting the IN2 population isdistributed across all layers of the EC. IN3 uniquely expressed LHX6 and CORT.Expression of CORT in the mouse adult brain is also found across all EC layerssuggestingdistributionofIN3throughout(Figure7E,G).TheIN4populationuniquelyexpressedGATA3 whilst the IN5 population expressed high levels ofDLX5 (Figure7E). GATA2 and 3 have been found in unique ventral IN subpopulations in thedeveloping chick, which supports the existence of this ventral population(Karunaratne et al., 2002). Similarly, DLX5 is an important regulator of PVALB-expression (Wang et al., 2010). The IN6 population was only detected duringembryonic gestation at E50 and E60 (Figure 7B), suggestive of a progenitor INpopulation. This population also clustered furthest from the remaining clusters(Figure7A)andexpressedhighlevelsofNDNF. WecoulddetectNDNF inLIandindeeperlayersintheE18.5mouseEC(Figure7G)highlightingtheputativelocationofthese INsduringdevelopment. In sum,we identified threeSST+and threeSST- INpopulationsintheEC,ofwhich1populationisonlypresentduringdevelopment.DiscussionWehaveclassifiedthestagingofneurogenesisinalargemammalianmodel,thepig,whichwehavefoundanexcellentchoiceforstudyingbraindevelopmentandwhichmirrorsthetemporaltimingofhumanneurogenesis.Givenitisextremelydifficulttoobtain human fetal tissue during the second and third trimester, the largegyrencephalic pig brain has proved to be a very good alternative model.NeurogenesisspansfromE39toE60intheventraltelencephalonofthepig(duringthe1st andearly 2nd trimester),which correspondsmore closely to the timingofneurogenesiswithinthehumancortex(gestationalweek7-17)(Kostovicetal.,2019)thantomice(E11-E17)(Casarosaetal.,1999;Royetal.,2004;Stagnietal.,2015).Theanatomicalorientationof thepigECwasalsosimilar tothat in thedevelopinghuman brain. To view the six-layered EC in the rodents, sagittal sectioning isrequired, whereas, we could better view the EC in the pig in the coronal plane,similar to the situation in the human. Immunohistochemical analyses acrossdevelopment resulted in thedetectionof amoderateOSVZduringE33-E60witha


https://doi.org/10.1101/738443

15

change in proportion of intermediate progenitors with mixed identity ofEOMES+/PAX6-andEOMES+/PAX6+toanenrichednumberofEOMES+/PAX6+cells.ConsideringthatPAX6isanimportanttranscriptionalregulatorofneurogenesiswithincreasing expression changing the balance from neural stem cell renewal toneurogenesis, (Sansom et al., 2009), we believe that neurogenesis lasts from E39until shortly after E60. We have discovered that the pig has a larger OSVZ thanrodents and that the timing of neurogenesis concurs with trimester matchedneurogenesisinthehuman.During gestation, entorhinal cells become visible already at E50 (during the earlysecond trimester) and the EC becomes multilayered by E60, comprising keycytoarchitectonicfeaturesdelineatingtheMECfromtheLEC.EarlyemergenceoftheEC in humans occurs towards the end of the first trimester with prominententorhinalcellsvisibleby10.5weekspostovulation(Kostovicetal.,1993),whichisthus closely aligned between these two species. In the rat, the entorhinal cellsemerge later, at E16 (Deng et al., 2006), and in themouse, stellate cells (RELN+)emerge by E12, with pyramidal neurons emerging approximately 2 days later(Donatoetal.,2017).Inourstudy,wefoundthesuperficialBCL11B+entorhinalcellsemerged at E50 and BCL11B+/RELN+ cells could be detected from E60 onwards,indicatingagap inbirthandexpressionofmaturingneuronswithin the superficiallayers.Together,ourstudyhighlightsthatthepigsharestemporalsimilaritiesinthedevelopingECtothehuman.TheECemergesinasandwichlaminationpatternwithLIIneuronsbornearlyduringneurogenesis.MaturationoftheMEChasbeendescribedrecentlyasdorsoventralinthemouse, which correlates with the birth of the stellate cells, but not with thepyramidal neurons (Donato et al., 2017). We extend this data to show that LIIsuperficial neurons (we do not discern between stellate and pyramidal) are bornprior to LIII as also reported in Donato et al., and that a sandwich patterning ofcortical laminationoccurs,whichhasnotbeenwellrecognized.Thispatterninghasalsobeenreported in theanatomicalclassificationof thedevelopingEC in therat,whichsuccinctlyshowsthatLIII formsafterLII (Bayeretal.,1993).Weconcurwiththis earlier study and supplement the research with labeling of superficial versusdeep layermarkers, in combinationwith a temporal analysis of expression acrossgestation. Our additional analyses show that this altered expression of markerspersistsuntil afterbirth.Bromodeoxyuridine labelingwould certainlyhelpuncoverthepreciseemergenceofthedifferent layers intheECanddeterminewhetherLIIIemerges prior to, or after LV and LIV. However, given the quantities of BrDUrequired for administration of >200 kg pregnant sowswe did not embark on thisstudy.Ofparticularinterest,ourdatatogetherwithISHdatafromAllenBrainAtlasindicatedivergingexpressionpatternsofBCL11Bbetweenmouseandpigbrain.Anincreaseofthetranscriptionalregulator,PAX6inapicalprogenitorsisimportantforthegenerationofsuperficialcorticalneurons(Georgalaetal.,2011).Therefore,itisplausiblethatPAX6expression levelsmightdiffer intheECversusotherregions. Itwould also be of interest to examinewhether altered RELN expressionwithin thelocalCRcellscouldaffectthelaminationpatterning.Furtherresearchisrequiredto


https://doi.org/10.1101/738443

16

investigate the signalingmechanisms underlying this unique cortical developmentfeature.We identified two OLIG2+ OPC populations in the LEC which reside in differentlayers.Thepan-neuronalclusteringandabundanceofOPCsintheLECconcurswiththe particular enrichment of projections we observed in the DTI from the LEC’ssuperficialprojectionneurons.TheearlyemergenceoftheseOPCsduringgestationsuggests that at least one wave of oligodendrogenesis overlaps with ECneurogenesis. In mice, the first OPCs emerge from the ventral medial ganglioniceminenceatE11.5andmigratethroughoutthetelencephaloninaventraltodorsalmannermidwaythroughgestation(Kessarisetal.,2006),whichisonlyslightlylaterthanwhatweobserveinthepig.OneOPCpopulation(OPC1)continuedtoconserveitsOPCsignatureuntilafterbirth,whereasanotherpopulation(OPC0)residinginthesuperficiallayerslostitsidentitylaterduringgestation,likelyduetotransitioningtoamorematurephenotype(potentiallyOLI2).TheremainingOPCpopulationresidingin the EC even after birth is an interesting population. Research has previouslyindicatedthatOPCscangiverisetopyramidalneuronsinthepiriformcortexintheadultmouse (Guo et al., 2010). Such plasticity has not been described in the EC,however itwouldbe interestingtotracethefateofthesecells intheadultECandevaluate EC plasticity. The pre-myelinating oligodendrocyte phenotype (OLI2)expressedMBP and emerged quite early during gestation, at E60,which indicatesmyelinationbeginsmuchearlierinthepigthanpreviouslythought.Apreviousstudyhasshownmyelinationofwhitefibersin2montholdpigs(Fangetal.,2005)butourstudy suggests themyelination processmay occurmuch earlier in the pig cortex.Together,wehighlightthepresenceoftwoOPCpopulationsthatresideinspatiallydifferentlocationswithinthedevelopingEC.At the single-cell level, we characterized the transcriptome of the developingentorhinalcortexcellsincludingtheexcitatoryneurons,inhibitoryneurons,gliaandprogenitors.Thedatasetprojectedextremelywellusingscmaptohumanandmousedatasets, which suggests this dataset can be usedwith confidence. Although, thehuman datasets were at unmatched gestational ages and from the embryonicprefrontalcortexandadultmiddletemporalgyrus,only2clusterswereunabletobeprojected to them. We were able to identify an IP population within the ECexpressing EOMES, SOX2 and NEUROD4 and 12 young/adult excitatory neuronprofiles,ofwhichinclude3youngpyramidaland3maturepyramidalneurontypes.Anadditional4populationsexpressingRELNhaddistinctprofilesofwhichtwowereexpressedonlyduringgestation,whilsttwootherswereexpressedevenafterbirth.Many of these clusters have been spatially mapped using their gene-specificsignatures to the adult EC. Some clusters exhibited amixed identitywith both anexcitatory and inhibitorymixed genotype.Wehave also discovered the speed celltranscriptional profile. This particular population was negative for vGLUT and SSTexpression,butexpressedRELNandGAD1/2.UniquemarkersexpressedintheSPE9cluster includedSFRP2,anotableWNTInhibitor(Kongkhametal.,2010) importantforaxonguidance(Lyuksyutovaetal.,2003),COL2A1,importantforchondrogenesis(Heringetal.,2014),COL18A1,importantineyedevelopment(Fukaietal.,2002),aswellasSIX3andSIX6,whichareimplicatedinretinalandanteriorbraindevelopment


https://doi.org/10.1101/738443

17

(Diacou et al., 2018; Lagutin et al., 2003). This unique gene set will be useful foridentifyingandstudyingthespeedcellinfutureresearch.Ourresearchdoeswarrantfollow up studies that can match the outlined transcriptomes more precisely tophenotypes, but the spatial dimension provided from tracing unique genes in themouse brain gives good indications of the cell types to which they belong. Forexample the identification of subcluster 10, expressed across gestation and in thepostnatal brain was the only RELN expressing cluster found postnatally andexclusively located inthesuperficial layer.Wethereforebelievethatthisgenotypebelongs to the Fan and/or stellate cell. Concurrently, cluster 3 has a pyramidalneuron identity which extended in the postnatal brain and also resides in thesuperficial layer. Very few genemarkers can currently be used to identify the celltypesoftheEC.Untilnow,onlyRELNandCALB1couldbeusedtodecipherstellatecellsfrompyramidalneurons(Perez-Garciaetal.,2001;Seressetal.,1994).Here,weprovide more encompassing gene signatures for a diverse number of detectableexcitatoryandinhibitoryneuroncellpopulations.Together, we bring forth a novel animal model for studying brain developmentwhich we have used to trace in detail the developing entorhinal cortex. Wedescribed a sandwich-structured EC with the LII forming prior to LIII. We havedecodedthemolecularidentityofthecelltypesintheECandusingspatialmappingof unique genes identified unique gene signatures that together with spatialpositioningofuniquemarkersfromthesesignatures,identifytheidentityofseveralcelltypesintheEC.ThisresearchhelpsfillthevoidofthegeneticconstitutionoftheEC cells and further research will be useful for connecting the genotypes to thephenotypes to gain even more precision in the understanding of the functionalcircuitryofthespatialnavigationalsystem.Acknowledgements:WethankPerTorpSangildforhisdonationofhealthyadultsowbrainsforthisstudy.WeacknowledgetheCoreFacilityforIntegratedMicroscopy,FacultyofHealthandMedicalSciences,UniversityofCopenhagenforaccessanduseoftheAxioScanZ.1.WealsoacknowledgeAnnaFossum,FACSFacilityatBRIC,UniversityofCopenhagenfor helpwith the sorting. The projectwas financed by The Independent ResearchFund,Denmarkunderthegrant (ID:DFF–7017-00071and8021-00048B),Lundbeckfund(R296-2018-2287)andThe InnovationFoundation(Brainstem(4108-00008B)),Novo Nordisk Hallas-Møller Investigator grant (NNF16OC0019920). THPacknowledgestheNovoNordiskFoundation(GrantnumberNNF18CC0034900)andthe Lundbeck Foundation (Grant number R190-2014-3904). MH and NP weresupported by a core grant to the Wellcome Sanger Institute from the WellcomeTrust.LFHwassupportedbyOpenTargets(OTAR038)andJLwassupportedbyaCZI(2018-183503(5022))fromtheChanZuckerbergInitiative.WealsoacknowledgetheSingle Cell Core Facility at BRIC, University of Copenhagen for access to sc-seqservices.


https://doi.org/10.1101/738443

18

AuthorContributions:Termsusedtodescribecontributionhavebeensentaccordingtotherecommendedlist:https://www.cell.com/pb/assets/raw/shared/guidelines/CRediT-taxonomy.pdfYL:Methodology; Investigation; FormalAnalysis;Writing –OriginalDraft,Writing –Review&Editing,VisualizationTB:Methodology; Investigation; FormalAnalysis;Writing–OriginalDraft,Writing–Review&Editing,Visualization

JL:Investigation

UP:Investigation

L-FH:Software

YM:Investigation,FormalAnalysis

AAM:Investigation

ILV;InvestigationSES:Investigation,SupervisionJTHL:SoftwareNP:SoftwareJMPV:Investigation,FormalAnalysis

MP:Investigation

BRK:Resources

PDT:Resources

PH:Resources,Supervision,FundingAcquisition

MPW:FormalAnalysis,FundingAcquisition

KK:Resources,Supervision,Methodology

JG:Resources,Supervision

MH:Software,Resources,Supervision

THP:Resources;FormalAnalysis;Software

VH: Conceptualization; Methodology; Validation; Formal Analysis; Resources,Writing–OriginalDraft,Writing–Review&Editing,Visualization,Supervision,ProjectAdministration,FundingAcquisitionDeclarationofInterests:Theauthorsdeclarenocompetinginterests.FiguretitlesandlegendsFigure1.Cytoarchitectonic featuresof thedevelopingporcineEC. (A) Lengthandtrimesterdivisionsofthegestationinmouse,human,andpig.(B)Macroscopicviewof theventralsurfaceof theporcinebrainatE100withannotationof thepiriformlobe including the entorhinal area. Olfactory tract (Olf.Tr.); caudate nucleus ofamygdala (COA); hippocampal area (HA); lateral entorhinal cortex (LEC); medialentorhinal cortex (MEC); sagittal sulcus (S.Sag); posterior rhinal sulcus (RHP). (C)Delineationsandcytoarchitectonicfeaturesofcresylvioletstainedcoronalsections


https://doi.org/10.1101/738443

19

of theEC (shownatE100as representative). Lightgreypunctuated linemarks theacellularL4 laminadissecans,blackpunctuatedlinebordersbetweenMECandLECor EC and adjacent areas. Red arrows (middlepanels) shows thedifference in thecellularorganizationindeeplayers,arrowheadshighlightlargesuperficialcellsoftheEC. Medial (M); lateral (L); pre-subiculum (PreS); para-subiculum (PAS), perirhinalcortex (PER); layer 2 - 5 (LII - LV). (D) Oligodendroglia-like cells in the superficiallayers of the EC and adjacent areas. Neocortex (NEO). High magnification of redsquares in (E) where the morphology of the oligodendroglia-like cells isdemonstratedatE100(redarrowheads).

Figure 2.Magnetic resonance imaging (MRI) of the developing EC and diffusiontensor imaging (DTI) of the white matter projections shows a spurt in growthduringlate2ndtrimesterandprojectionstovaryingbrainregionsatP75.(A)MRIshowstherostraltocaudalpositionoftheearlydevelopingECandpostnatalEC inthecoronalplane(markedinorange).(B)TheECvolumeshowsaparticulargrowthspurtatE70whencomparedwithcortexvolumechangesduringdevelopment. (C)AnoverviewofwhitematterfibersconnectingtotheMEC(orange)andLEC(purple).(D) The average number of tracts connecting to the MEC and LEC do not differsignificantly. (E)Whitematter fibers connecting to theperirhinal cortex (triangles)and postrhinal cortex (arrows). (F)Whitematter fibers extend from the EC to thecorpuscallosum(arrow). (G)Whitematter fibersconnectingto the lateral (arrow)and medial olfactory bulb (triangle). (H) White matter fibers connecting to thehippocampus(arrow).(I)Whitematterfibersconnectingtotheamygdala(triangle).(J) Average number of tracts connecting to different anatomical regions emergingfrom the EC do not significantly differ between the MEC and LEC. All error barsdenoteSD.n.s.p>0.05,*P≤0.05,**P≤0.01,***P≤0.001.Figure3.AlteredexpressionofsuperficialanddeeplayermarkersintheECpersistsuntilafterbirth.(A)Locationofanalyzedventraltelencephalon(V),MEC,anddorsaltelencephalon (D) at E50 and E60 of development. (B) Altered expression ofsuperficial layer marker SATB2 and deep layer marker BCL11B in the ventraltelencephalon and MEC combined with the expression of RELN compared to thedorsal telencephalon. (C)ExpressionofSATB2,BCL11B,andRELN in theMECfromE70untilP75ofdevelopment.(D)LocationofanalyzedsuperficiallayersacrosstheEC,includingtheMECandLEC.(E)RELN+neuronsco-expressBCL11BacrosstheECduringgestationandalsoafterbirth.Scalebar50μm(B,C)and25μm(E).Figure4.Single-cellprofilingoftheECreveals32distinctcellclustersofmajorcelltypepopulations in the EC. (A)Thebioinformatic pipeline includes 10batchesofthewhole cell and isolated nuclei cells from E50, E60, E70 and Adultwhichwerecapturedassinglecellsusingthe10xgenomicsplatform.BatcheswerenormalizedusingtheFastMNNapproach,dimensionalreductionandclusteringwereperformedinSeuratandsubclusteringandanalysesofselectedclusterswasperformed.(B)At-SNEplotof24,294cellsmergedfromalltimepointsrevealed32distinctpopulations.(C) Analysis of canonical marker genes resulted in categorization of severaloligodendro-glia/-cytes (Oligo), astrocytes and radial glia (Glial cells), intermediate


https://doi.org/10.1101/738443

20

progenitors (IP), excitatory neurons (Excitatory), interneurons (IN) and microglia(Microglia) clusters, together with the number of cells, mean UMI counts andexpressed genes for each cluster. Error bars denote SD. (D) Batch correction byFastMNNresultedsatisfactorymergeofthe10datasets(pre-batchcorrectionshownin Figure S5C). (E) Projection of the dataset onto already-annotated human fetalprefrontalcortexandhumanmedialtemporalgyrussingle-celldatasetconsolidatesmarkergene-drivenannotationofthedataset.BloodcellsexpressingHBB.Figure 5. Oligodendrocytes and precursors in the EC. (A) Visualization ofoligodendrogliasubset(2,696cells)in5distinctclusterinat-SNEplot.Subsetoftheoligodendroglia(orange)withintheparentdatasetisdepictedintheinsertinthetopleft corner. (B) Developmental stages of the oligodendroglia visualized in a t-SNEplot. (C) Canonical lineagemarkers and unique gene signatures for the 5 clustersidentifiedbydifferentialgeneexpressionanalysis.Localizationofgeneswithnamesinblueisshownin(E).*humanortholognames(Supp.Table2).(D)Distributionofcells in different cell-cycle phases in the 5 clusters. Notably, G1-phase isindistinguishablefromG0-phaseofcell-cycle.(E)SagittalsectionsofmurineEC(fromAllenBrainAtlas)withISHstaining,ofeitherfetal(E18.5)oradultbrainsections.Figure 6. Molecular diversity of the excitatory neurons and intermediateprogenitors of the EC. (A) A t-SNE plot of excitatory neurons and intermediateprogenitors (7,627 cells) in 13 distinct clusters. Excitatory neurons (blue) andintermediateprogenitors(green)whichweresubsettedaredepictedininsertoftheparentdataset. (B)Distributionofdevelopmentalstageswithin the13clusters. (C)Uniquegenesignaturesforthe13clusters.Localizationofgeneswithnamesinblueis shown in (D). * human ortholog names (Supp. Table 2). (D) Sagittal sections ofmurineEC(fromAllenBrainAtlas)withISHstaining,ofeitherfetal(E18.5)oradultbrainsections.Figure 7. Subclustering of EC interneurons reveals 6 clusters with unique genesignatures.(A)At-SNEplotofinterneurons(IN)forming7clusters.(B)Distributionof developmental stages within the 7 clusters. (C) Expression of canonical INmarkers. (D) Unique gene signatures identified for the 7 clusters. Localization ofgeneswithnamesinblueisshownin(F).*humanortholognames(Supp.Table2).(E)SSTandBCL11BexpressionintheMEC.CoronalsectionofthepigECatE60.Scalebar 50μm (left) and 25μm (right). (F) Sagittal sections ofmouse E18.5 and adultEC(fromAllenBrainAtlas,withISHstaining.Scalebar200μm.(F)SSTandBCL11Bexpression intheMEC.CoronalsectionofthepigECatE60.Scalebar50μm(left)and25μm(right).


https://doi.org/10.1101/738443

21

STARMETHODSCONTACTFORREAGENTANDRESOURCESHARINGFurtherinformationandrequestsforresourcesandreagentsmaybedirectedtoandwillbefulfilledbytheLeadContact,VanessaHall([email protected]).EXPERIMENTALMODELANDSUBJECTDETAILSAnimalwelfareandcollectionofbrainsThe experiments conform to the relevant regulatory standards for use of fetalmaterial.CrossbredmaleandfemalepigfetusesatE26,E33,E39,E50,E60,E70,E80and E100 of development were obtained from inseminated DanishLandrace/YorkshiresowswithDurocboarsemenfroma localpigproduction farm.Note, theaveragegestation length for this crossbreed is 114days. Theuteriwereremovedfollowingtheslaughterofthesowsbytrainedandlicensedpersonnelatalocal slaughterhouse and transported warm at 37°C to the University where thefetuses (clinically dead from asphyxiation during transport) were then isolated byumbilical excision from their fetal sacs. Deceased postnatal pigswere obtained atP75 as a gift fromPer Torp Sangild at theUniversity of Copenhagen. Adult brainswere obtained from sows killed for another study using an overdose of sodiumphenobarbital by a professional issued with a license from the Danish AnimalExperimentInspectorate.METHODDETAILSBrainFixationandstorageFor the earlier time points up until E60, the entire fetuswas fixed and the dorsalskullwasopened to improvepermeationof the fixative. For the later timepoints,the brains were removed from the skull. Fixation was performed using 4 % PFA(Millipore Sigma) in PBS (Thermo Fisher Scientific) from 24 hrs to up to 2 weeks,dependentonthesizeofthefetus/brain.Fetusesandbrainswerethenstoredlong-termat4°Cin0.002%SodiumAzide(VWR-Bie＆Berntsen)inPBS.ParaffinembeddingandsectioningPriortodehydration,thebrainsweredissectedtoapproximately10mmthickness.Thebrainsweredehydratedbyimmersionintoasequentialseriesofethanol,estisoland liquidparaffinusinga tissueProcessor (ThermoFisherScientificCitadel2000).Following dehydration, the tissues were embedded in liquid paraffin followed bycooling down on a cold stage. Five-micrometer (μM) thick sectionswere cut on amicrotome (Leica SM2000R) and mounted onto SuperFrost slides (Thermo FisherScientific). The sectionswere dried at room temperature (RT) overnight (ON) andstoredlongtermat4°C.


https://doi.org/10.1101/738443

22

CresylVioletStainingParaffinembeddedsectionsweredeparaffinizedfor45minat60oCand2x10mininXylene.Thesectionsweresequentiallyrehydratedfor2x5minin99%EtOH,3minin96%EtOH,3min in70%EtOH, rinsed in tapwaterandstained for12min in37oCCresyl Violet (Millipore Sigma). The stained sectionswere rinsed in distilledwaterand staining outside of perikaryawas differentiated for 10min in 96% EtOHwith0.02% glacial acetic acid (Millipore Sigma). The sections were subsequentlydehydrated2x5minin99%EtOHand2x5mininXylenepriortocoverslipmountingusingDPX(Sigma-Aldrich).AllimageswereacquiredonanAxioScan.Z1(Zeiss)usingautomatedbrightfieldimaging.Tissue-TekOCTembeddingandCryosectioningFixed fetuses/brains were dehydrated in 30% sucrose(Sigma-Aldrich)/1x PBS(Thermo Fisher Scientific) solution (0.22 μm filtered) at 4°C for 48 hours (h). Thebrainswerecutintosmallpiecesofapproximately4mmthickness.Thebraintissueswere immersed into Tissue-Tek OCT (Sakura) and mounted within plastic molds(Simport) by snap-freezing in N-hexane ((VWR) solution immersed within liquidnitrogen.Thefrozentissuewasstoredat-80°Cuntiluse.Brainsectionswerecutat30μmthicknessusingacryostat(LeicaCM1950)andmountedontoSuperFrostPlusslides(ThermoFisherScientific)andstoredat-20°C.ImmunohistochemistryParaffinembeddedbrainsectionsweredeparaffinizedfor45minat60oCintheovenand 2x 10 min in Xylene(VWR). Subsequently, the sections were sequentiallyrehydratedfor2x5minin99%EtOH,3minin96%EtOH,3minin70%EtOH,3minin tap water, and washed 2x for 5 min in PBS. The sections were subsequentlypermeabilizedfor30minin0.1%Triton-X(MilliporeSigma)inPBSatRTandwashed2x 5min in PBS before antigen retrieval in boiling citrate [0.01M, pH6] (Sigma-Aldrich) for 3x 5 min. The sections were washed once for 5 min in PBS and LabVision™MultiVisionPolymerDetectionSystem (ThermoFisherScientific)wasusedaccording to manufacturer’s protocol with the following specifications: Anti-parvalbumin (Millipore Sigma) primary antibody was diluted in 1:500 in PBS andincubatedatRTfor30min.ThesectionswerevisualizedwithLVRedandLVBlue.ImmunofluorescenceCryosectioned brainswere air-dried for 1hr at RT, followed by rehydration in PBS(Thermo Fisher Scientific) for 10 min. After the antigen retrieval (detaileddescription at Immunohistochemistry staining) the sectionswere permeabilized in0.25%TritonX-100/PBS(Sigma-Aldrich)for10minutesfollowedbywashinginPBS.Thebraintissuewasdemarcatedwithahydrophobicpenfollowedbyblockinginthe


https://doi.org/10.1101/738443

23

blockingbufferwhichcontain10%Normaldonkeyserum(Biowest)2%BSA(Sigma-Aldrich) at RT for 1 h. The sectionswere then incubatedwith the diluted primaryantibody(seebelow)ONat4°CfollowedbywashinginPBS3timesfor5minseach.The sectionswere then incubated in diluted secondary antibody for 2h at RT andthereafter washed in PBS for 3 times for 5 mins each. The sections werecounterstainedwith10μg/mlHoechst33258(Sigma-Aldrich)for10minfollowedbywashing in PBS for 5mins twice. Finally, the sectionsweremountedwith a glasscoverslip in fluorescence mounting medium (DAKO). The primary antibodies andsecondaryantibodiesweredilutedwithblockingbuffer.Theprimaryantibodiesusedareasthesourcetable.Differentantibodiescombinationswereappliedtothesamesections during individual reactions according to the proteins under study.Secondaryantibodiesweredilutedat1:200andselectedtobethesamespeciesasthe species used for producing the primary antibodies and were conjugated witheitherAlexafluorophores,488,594or647.ImageprocessingandquantificationFor acquisition of Cresyl Violet images, a bright-fieldmicroscope, Leica DMRwithcameraLeicaDFC490andZeissAxioscan.Z1wereusedtocapturethe images.Foracquisitionof immunohistochemistry images, a confocalmicroscope Leica TCS SPEwasused.Samplesbelonging to thesameexperiment (samples fromexperimentalpigatagiventimepoint,withtheircontrols)wereacquiredinparallelandwiththesame settings (laser power: 8-20%; optical slice: 1 airy units, step size: 3 µm forpopulationanalysis)using40×/1.15oilimmersionobjective.Priortotheacquisition,gainanddigitaloffsetwereestablishedonsectionsfromsecondaryantibodycontrolsample to optimize the dynamic range of acquisition to the dynamic range of thestaining (baselineswere set independently forevery stainingbasedon theproteinunder investigation). Settings were kept constant during acquisition. ConfocalimageswereacquiredusingLeicaLASX. Imageswereoptimizedforbrightnessandcontrast using Fiji-ImageJ. Statistical analysis was conducted with commerciallyavailablesoftware-Prism7.0(GraphPadSoftware).StructuralMRIA 9.4T BioSpec 94/30 USR spectrometer (Bruker BioSpin, Ettlingen, Germany)equippedwitha240mT/mgradientsystemwasusedtoacquireanatomicalimagesofpostmortemporcinebrains(E60,E70,E80,E100,P75).Priortoimaging,thePFAfixed brain were suspended into plastic containers filled with a proton-freeperfluorinatedsusceptibility-matchingfluid(SolvayGalden®HT-230).Weuseda35-mm (for P75) and a 23-mm (for E60, E70, E80, and E100) inner diametertransmit/receive volumecoil (Bruker). TheMR systemwas interfaced toa consolerunningParaVisionsoftware6.0.1(BrukerBioSpin).Theparametersusedinthebrainscans were optimized for gray/white matter contrast: T2-weighted 2D rapidacquisition with relaxation enhancement (RARE) pulse sequence with TR/TE =20000/40 ms (before birth), 12000/60 ms (after birth), Rare factor = 8 (beforebirth),16(afterbirth),in-planeresolution=100µm×100µm,slicethickness=200


https://doi.org/10.1101/738443

24

μm (before birth), 500μm (after birth). 3D fast low angle shot (3D-FLASH) pulsesequencewithTR/TE=30/4.6ms,Flipangle=10,NA=16,spatial resolution=47µm×47µm×70µm.Thetotalimagingtimewas2hours.DiffusionMRItractographyEx vivo diffusion tensor imaging (DTI) of P75 brain (n = 3) was obtained using aStejskal-Tannersequence(TR/TE=3500/17.5ms,NA=2,spatialresolution=390µm×390µm×390µm)ona9.4Tspectrometerequippedwitha1500mT/mgradientsystemanda40-mm innerdiameter transmit/receivevolumecoil.Weapplied themotionprobinggradients(MPGs)in60noncollineardirectionswithb=2500s/mm2in addition to 4 b0. Diffusion Toolkit (version 0.6.4.1) and TrackVis (version 0.6.1)were used for 3D reconstruction of white matter tracts. We used a DWI maskthreshold and an angular threshold of 45 degrees. Two regions of interest (ROIs)were selected a priori for theDTI analysis: including the LEC andMEC. ROIswerecreatedbymanually segmentationwith ITK-snap (version3.6.0, (Yushkevichet al.,2006)).ThefibertracksobtainedfromtheROIswerequalitativelyanalyzedforthedirectionof tracks, anddifferences in FAvalueswithin thebrainparenchyma.Theassessment was done using tractography maps: (1) a standard color-codedreconstruction tovisualize theorientationof trackswhereblue tracks representeddiffusion in the craniocaudal direction, green tracks represented diffusion in theanterior-posterior direction, and red tracks showed diffusion in the transversedirection;(2)ascalarFAmapwithminimumandmaximumFAthresholdsof0.1and0.6,respectively;and(3)totaltracklengthinformationforeachROI.

Single-cellpreparationIn totalwe prepared 10 sequencing-libraries form isolatedMECs from E50 (wholecell,3brains,1batch),E60(wholecell,4brains,3batches),E70(wholecell,4brains,3batches;nucleus1brain,onebatch)andfromadultsowMEC(wholecell,1brain,1batch;nucleus,1brain,1batch)(TableS1).Briefly,theindividualMECtissuewasdigested using a papain dissociation method, according to the manufacturer’sguidelines(Worthington)withsmallmodifications.TheECwasmacroscopicallydissectedout(approx.1mm3)inthedigestmedium(1xPBS (Thermo Fisher Scientific), 1x Penicillin-Streptomycin (Sigma-Aldrich)),transferredtoa3.5cmPetridishandincubatedin1mLpapainsolutionfor30minat37oC.Solutionandremainingtissueweresubsequentlytransferredtoa15mLfalconand gently triturated 20 times. The cell suspension was diluted with 1 mL FBS(BioWest)andcentrifugedfor5minat300g,RT.Thesupernatantwasdiscardedandthecellpelletwasresuspendedinpreparedsolutionwith2.7mLdigestionmedia(1xNeurobasal medium (Thermo Fisher Scientific), 10% FBS (BioWest), 1x Penicillin-Streptomycin (Sigma-Aldrich)), 300 uL albumin-ovomucoid inhibitor, and 150 uLDNAse solution. The cell suspension was carefully layered on top of 5.0 ml ofalbumin-inhibitorsolutionina15mLfalcontubeandcentrifugedfor6minat70g,RT.Thesupernatantwasdiscardedandthecellpelletwasresuspendedin5mLcell-


https://doi.org/10.1101/738443

25

resuspensionmedia(1xNeurobasalmedium(ThermoFisherScientific),B27(ThermoFisher Scientific), 1x Penicillin-Streptomycin (Sigma-Aldrich), bFGF (5 ng/ml,Prospec).Thecellswerecounted(NucleoCounter,ChemoMetec)anddilutedto100-2000cells/uLusedforsingle-celllibrarypreparation.SinglenucleiisolationNuclei extraction was performed as described before with the followingmodifications(Krishnaswamietal.,2016).Priortonucleiextraction,nucleiisolationmedium1 (NIM1) (250mMsucrose,25mMKCl, 5mMMgCl2,10mMTrisBufferpH8),NIM2 (NIM1buffer supplementedwith 1 μMDTT (Thermo Fisher Scientific)and 1x EDTA-free protease inhibitors (Roche) and homogenization (NIM2 buffersupplemented with Recombinant RNase Inhibitor (0.4 U/μL, Takara), SUPERase in(0.2 U/μL, Thermo Fisher Scientific) and Triton (0.1% v/v, Sigma-Aldrich)) bufferswereprepared.Briefly,sectionedfrozenbraintissuewasplacedintopre-cooled1mldouncehomogenizer(Wheaton)with1mlice-cooledhomogenizationbuffer.Tissuewasdissociatedoniceusing5-6strokeswiththeloosepestleand15-17strokeswiththetightpestle.Homogenatewas first filteredthrougha70μmfilter.Nucleiwerecollected (900 g, 10min) and resuspended in 500μl staining buffer (nuclease freePBS (1X, Thermo Fisher Scientific), BSA (0.5% wt/vol, Sigma-Aldrich), SUPERase in(0.2 U/μL, Thermo Fisher Scientific)). Then the sample was stained by the 7-AAD(2μg/ml,Sigma-Aldrich)followedbyFACSsorting(70μmnozzle,BDBiosciences,BDFACSAria™) to 1.5ml eppendorf tube containedwith 10 μl 10% nuclease free BSA(ThermoFisherScientific).SingleCell/NucleiRNA-seqlibrarypreparationandsequencingThewholecellswere loadedontothe10XGenomicsmicrofluidicchipaccordingtothe Chromium Single Cell 3' Reagent Kits User Guide (10X Genomics) followingadequate dilution. The single nuclei sampleswere loaded onto the 10XGenomicsmicrofluidicchipsimilarly,albeitthesampleswerenotdiluted.10-12.000thousandcells/nucleiwereloadedfromeachsample.LibrariesfromtwosamplesatdifferentstageswerepooledandsequencedtogetheronanIlluminaNextSeq500(TableS1)followingtheNextSeqSystemDenatureandDilute Libraries Guide Protocol A: Standard Normalization Method (illumina). TheNextSeq500/550HighOutputReagentCartridgev275cycles(illumina)kitwasusedforthewholecellandsinglenucleisamplesandthepooledlibraryweresequencedonNextSeq500.Thesequencingcycleswere:Read1,26cycles,i7Index8cycles,i5Index0cycles;Read2,57cycles.


https://doi.org/10.1101/738443

26

QUANTIFICATIONANDSTATISTICALANALYSISStatistics:For immunohistochemistry and assessement of neuroconnectivity from DTI whitemattertracts,allsampleswereperformedinbiologicaltriplicates.Statisticalanalysiswas conducted using commercially available software- Prism 7.0 (GraphPadSoftware).Dependentontheexperimentaldesign,anordinaryone-wayANOVAoran unpaired two-tailed t testwas performed to statistically assess differences. AllerrorbarsdenoteSD.n.s.p>0.05,*P≤0.05,**P≤0.01,***P≤0.001.InitialqualitycontrolanddataanalysisBriefly, the sequenced librariesweremapped to pre-mRNA and filtered using 10XGenomicsCellRangerpipeline.Thesequencingdatawasdemultiplexedbybcl2fast (illumina)whichwarped intheCell Ranger followed by aligning to reference genome (Sscrofa11.1 release-94) bySTAR(Dobinetal.,2013).Finally,mRNAmoleculeswerecountedandbyCellRanger.The quality of the sequencing libraries was assessed by Cell Ranger, whichdeterminedineachsamplethesequencingdepthcutoffthatisrequiredforcellstobe included in the downstream analysis. Similar number of UMIs and geneswereobservedacrossallbatches(FigureS5D-E).BatchcorrectionWeuseddecomposeVarfromthe‘scran’Rpackage(RRID:SCR_001905)inordertofindalistofvariablegenesthatareusedforPCAdimensionalityreduction;however,toaidthecorrespondenceofsinglecelltosinglenucleus,weidentifiedandexcludedgenes whose transcripts were highly abundant in empty droplets (cells with <50UMIs). If thenumbertranscriptsthatarefound intheemptydroplet isabove30%the total number of transcripts for a given gene (Figure S5F), that gene will beremovedfromthelistofvariablegenes.TheprojectedgeneexpressionwasthenbatchcorrectedbyFastMNN(Haghverdietal., 2018), which reduced the distance between cell pairs that are found to bereciprocallynearestneighborsacrossbatches (beforebatchcorrection;FigureS5C,after correction; Figure 4B). The merged dataset were visualized by t-distributedstochasticneighbourembedding(t-SNE)(vanderMaatenandHinton,2008).UnsupervisedclusteringandcelltypeidentificationCell typeswere identifiedusing theLouvainalgorithmwitha resolutionparametersetat1.6forallclusteringanalyses.Thecanonicalmarkerswereusedtoidentifytheneurons of the clusters (Figure 4C). In addition, we used two reference datasets


https://doi.org/10.1101/738443

27

which each contained smart-seq2 single-cells from human embryonic prefrontalcortex (Fanetal.,2018;Zhongetal.,2018)or fromhumanadultmiddle temporalgyrus (Astick and Vanderhaeghen, 2018).We used Scmap (Kiselev et al., 2018) toproject individual cells onto curated cell-type clusters that are available in eachreference. Each cell-type prediction utilize the consensus of 3 similaritymeasuresfromqueriedcelltoreferenceclustercentroidsusingsetsofcell-typemarkersthatwereidentifiedintherespectivereferencedatasets;however,onlythehumangenesthat possess an ortholog gene in pig were used as cell typemarker for similaritymeasurecalculation.DifferentialgeneexpressionanalysisFor cluster cj, gene gi is considered a true positive (TP) if it is expressed, a falsenegative (FN) if it is not expressed, a false positive (FP) if is expressed in a cellassignedtoanothercluster,anda truenegative (TN) if it isnotexpressed inacellassignedtoadifferentcluster.Foreachgiweevaluateprecision=TP/(TP+FP),recall=TP/(TP+FN),andF1=2*precision*recall/(precision+recall).Foreachcj,genesarerankedbyF1withthehighestscoringgenesusedasmarkers.For analysis of enriched genes in the oligodendroglia population (Figure 5C) thesewerefoundusingSeuratsfunctionFindAllMarkers,withwilcoxonranksumtestusedand only considering genes expressed in at least 25% of the cells in either of thetestedpopulations.The5mosthighlyenrichedgenesarereportedhere.Cell-cyclescoreanalysisCell-cyclephaseswerescoredusingthebuilt infunctioninSeuratCellCycleScoring,whichwasusedtodeterminewhetheragivencellwaslikelytobeineitherS,G2MorG1(whichisindistinguishablefromG0)phaseofcell-cycle.Cell-cyclescoreswerebasedonalistofcell-cyclephase-specificgenesproposedbyTiroshetal.(Tiroshetal.,2016).DATAANDSOFTWAREAVAILABILITYDataavailabilityThedataispublicallyavailableatNCBIwithGEOaccessionnumberGSE134482.Thesubmitted data includes the raw sequencing data as fastq files together with theprocessed countmatrixused in this study. Thedata is alsouploaded toMendeleyhttps://data.mendeley.com/datasets/5fgznjfrzn/draft?a=071f733a-d5e4-424f-a211-d2e9467435bfCodeavailabilitygithub


https://doi.org/10.1101/738443

28

SUPPLEMENTALINFORMATIONSupplemental Information includessix figures, threetables,andonevideoandcanbefoundwiththisarticleonlineat..SUPPLEMENTALINFORMATIONTITLESANDLEGENDSFigure S1. Borders of the developing porcine entorhinal cortex (EC). Cresyl violetstained coronal sections of the piriform lobe from E60 to P75. The first sectionrostraltotheEC,thesecondsectionwithLECoccupyingtheECentity,thirdsectionbothMECandLECpresent,fourthsectionMECoccupiesentiremediolateralentity,fifthsectionverycaudalpartofthepiriformlobe.Dentategyrus(DG);hippocampalarea(HA);amygdala(Amyg);posteriorrhinalsulcus(RHP);medialentorhinalcortex(MEC); lateral entorhinal cortex (LEC); pre-subiculum (PreS); para-subiculum (PAS),subiculum(Sub);perirhinalcortex(PER).Scalebar1cm.FigureS2.ValidationofdelineationoftheECbyparvalbumin(PVALB)stainingandlocationandexpressionofOLIG2oligodendrocyteprogenitorcells.(A)Cresylviolet(fromFigure S1) andPVALBexpression in the caudal EC at E100 andP75 (coronalplane).(B)Cresylvioletstainingofglia-likecellsatE50toE70,visibleassmallrounddarkstainedcells(C)CoronalsectionofE50brainthroughtheventraltelencephalondestinedtobecometheEC.Highmagnificationoftheventraltelencephalon(redbox)shows large superficial cells (arrowheads). (D)OLIG2 expression in the developingEC. Images taken in thedevelopingcorticalplate (E50-E60),and later (E70-P75), inthe superficial layers. (E) Quantification of OLIG2 positive cells in the EC bylayers/zones.Corticalplate,CP;intermediatezone,IZ;marginalzone,MZ;subplate,SP;subventricularzone,SVZventricularzone,VZ.AllerrorbarsrepresentSD.FigureS3.Characterizationof thegerminal layer in theporcineEC. (A)Schematicoverview of the location characterized. (B) Expression of GFAP and FABP7 in theventricularzone(VZ).Scalebar25μm.(C)Temporalexpressionofradialglia(GFAP,FABP7, PAX6, SOX2) during gestation. Scale bar 50 μm (D).Quantification of thethicknessinμmoftheVZduringgestation.(E)ExpressionofEOMESandPAX6intheEC. Scale bar 25 μm (up) and 50μm (bottom). (F) Quantification of theEOMES+/PAX6+ and EOMES+/PAX6- cell populations in the germinal zone. (G)TBR1/EOMES expression during gestation. Scale bar 50 μm. (H) Expression ofTBR1/EOMESinthemarginalzoneandthecorticalplateof(F).Scalebar25μm.(HO=Hoeschst,V=ventraltelencephalon).ErrorbarsrepresentSD.FigureS4.ComparativeexpressionofsuperficialanddeeplayermarkersintheECversus the dorsal, cingulate gyrus.Expression of the canonical deep layermarker(BCL11B),superficial layermarker (SATB2)andstellatecell /Cajal–Retziuscells (CRcells)byexpressionofRELNintheECandrhinalcortexofthetelencephalon.Scalebar50μm.


https://doi.org/10.1101/738443

29

Figure S5. Validation of single-cell sequencing data. (A) Top 10 gene features ofcluster32foundbyseurat(FindAllMarkerswithwilcoxon)fromthescRNA-seqdata.(B)Cell-cycle analysis of IP and excitatory neuron subpopulations (from Figure 5).Thecell-cyclescorewasassessedbasedontheexpressionofcell-cyclephase-specificgenes.G1phaseofcell-cycleintheanalysisisindistinguishablefromG0.(C)At-SNEplotofthesc-seqdatasetpriortofastMNNbatchcorrection.(D)ViolinplotofthenumberofUMIsandgenes ineach single-cell sequencingbatch. (E)A scatterplotnumbersofUMIsversusgenesforeachbatch.(F)Fractionoftranscriptsoutoftotaltranscripts for a given gene in empty droplets at representative stages (E70 andAdult),forbothisolatednucleiandwholecellsamples.Geneswithafractionabove0.30isnotincludedasvariablegenesindownstreamanalysis.

FigureS6.ProjectionofamouseP2andP11whole-braindatasetresultedinahighdegree of matching of cell types on the excitatory neurons and intermediateprogenitors of the EC. (A) t-SNE plot of the excitatory neurons and intermediateprogenitors of the EC after projection. Original clustering (from Figure 6A)representedbydottedlines.(B)Distributionofprojectedcell-typesintheexcitatoryneuronsandintermediateprogenitorsclusters.Top3cell-typesineachclusterwerelabeled with corresponding numbers to the projected data set, see panel A forlegend.SupplementaryTablesSupplementaryTable1.Connectivityofneuronaltractsto/fromthelateralandmedialentorhinalcortex.

Total Tracts MEC LEC AM CA1 CA3 DG PUT SUB LOT Caudate

nucleus Nucleus accumbens CC Other

MEC % 6205.0 - 106.0

(1.71) 730.3(11.77) 91.3

(1.47) 49.3(0.79) 47.0

(0.76) 4.7(0.08) 534.0

(8.61) 3.3(0.05) 9.0(0.15) 1.0

(0.02) 11.7(0.19) 4617.3

(74.41)

LEC % 4516.3 106.0

(2.35) - 468.7(10.38) 5.3

(0.12) 10.0(0.22) 2.7

(0.06) 59.0(1.31) 1.0

(0.02) 14.3(0.32) 75.3

(1.67) 0 (0) 0

(0) 3774.0(83.56)

Averageisshown(n=3)andstandarddeviationinbrackets. AM,amygdala,CC,Corpuscollosum;DG,dentategyrus;LEC,lateralentorhinalcortex;LOT,lateralolfactorytract;MEC,medialentorhinalcortex;PUT,putamen;SUB,subiculum

SupplementaryTable2.HumanorthologgenenamesidentifiedfrompigensemblenamesusingGeneCard(www.genecards.org).

EnsembleName Humanortholognames

ENSSSCG00000010212 ANK3

ENSSSCG00000011178 CPNE4


https://doi.org/10.1101/738443

30

ENSSSCG00000015545 GLUL

ENSSSCG00000009327 HMGB1

ENSSSCG00000035712 KCTD8

ENSSSCG00000011267 MOBP

ENSSSCG00000015401 PCLO

ENSSSCG00000007644 PILRB

ENSSSCG00000016219 RESP18

ENSSSCG00000033734 TMSB4X

ENSSSCG00000037477 TTC37

ENSSSCG00000029160 UNPROT1









ENSSSCG00000036465 USP6NL

ENSSSCG00000005658 ZDHHCL2

SupplementaryTable3.Overviewofbatches,capturedcells,meanreads,numbersoflibrariesandnumbersofbrainsforsc-seqexperiments.

Batch EstimatedNumberofCells

MeanReadsperCell

MedianGenesperCell

NumberofReads Library

Brainnumbers

E50_1 5506 49042 3448 270029135 2 3

E60_1 8171 27580 2129 225357492 3 1

E60_2 3479 68619 2548 238726637 4 1

E60_3 2696 98734 3335 266189304 1 2

E70_nc 858 305266 1584 261918893 5 1

E70_1 493 599954 4125 295777496 3 1

E70_2 617 484686 3927 299051391 2 1

E70_3 3120 80781 2990 252038762 1 2

Adult_nc 2552 90774 560 231655848 5 1

Adult_1 737 389148 3214 286802165 4 1


https://doi.org/10.1101/738443

31

Total 28229

ReferencesArtegiani, B., Lyubimova, A., Muraro, M., van Es, J.H., van Oudenaarden, A., andClevers, H. (2017). A Single-Cell RNA Sequencing Study Reveals Cellular andMolecularDynamicsoftheHippocampalNeurogenicNiche.CellRep21,3271-3284.Astick,M., and Vanderhaeghen, P. (2018). FromHuman Pluripotent Stem Cells toCorticalCircuits.CurrTopDevBiol129,67-98.Bayer,S.A.,Altman,J.,Russo,R.J.,andZhang,X.(1993).Timetablesofneurogenesisin the human brain based on experimentally determined patterns in the rat.Neurotoxicology14,83-144.Beall, M.J., and Lewis, D.A. (1992). Heterogeneity of layer II neurons in humanentorhinalcortex.JCompNeurol321,241-266.Beed, P., Gundlfinger, A., Schneiderbauer, S., Song, J., Bohm, C., Burgalossi, A.,Brecht, M., Vida, I., and Schmitz, D. (2013). Inhibitory gradient along thedorsoventralaxisinthemedialentorhinalcortex.Neuron79,1197-1207.Bergmann, E., Zur, G., Bershadsky, G., and Kahn, I. (2016). The Organization ofMouseandHumanCortico-HippocampalNetworksEstimatedbyIntrinsicFunctionalConnectivity.CerebCortex26,4497-4512.Bjarkam,C.R.,Glud,A.N.,Orlowski,D.,Sorensen,J.C.H.,andPalomero-Gallagher,N.(2017). The telencephalon of the Gottingen minipig, cytoarchitecture and corticalsurfaceanatomy.BrainStructFunct222,2093-2114.Bjerknes, T.L., Langston, R.F., Kruge, I.U., Moser, E.I., and Moser, M.B. (2015).Coherenceamongheaddirectioncellsbeforeeyeopeninginratpups.CurrBiol25,103-108.Bjerknes, T.L., Moser, E.I., and Moser, M.B. (2014). Representation of geometricbordersinthedevelopingrat.Neuron82,71-78.Burwell,R.D.,andAmaral,D.G.(1998).Perirhinalandpostrhinalcorticesoftherat:interconnectivity and connectionswith the entorhinal cortex. J CompNeurol 391,293-321.Canto,C.B.,andWitter,M.P.(2012a).Cellularpropertiesofprincipalneuronsintheratentorhinalcortex.I.Thelateralentorhinalcortex.Hippocampus22,1256-1276.Canto,C.B.,andWitter,M.P.(2012b).Cellularpropertiesofprincipalneuronsintheratentorhinalcortex.II.Themedialentorhinalcortex.Hippocampus22,1277-1299.Canto, C.B.,Wouterlood, F.G., andWitter,M.P. (2008).Whatdoes the anatomicalorganizationoftheentorhinalcortextellus?NeuralPlast2008,381243.Casarosa,S.,Fode,C.,andGuillemot,F.(1999).Mash1regulatesneurogenesisintheventraltelencephalon.Development126,525-534.Chan, C.H., Godinho, L.N., Thomaidou, D., Tan, S.S., Gulisano,M., and Parnavelas,J.G. (2001). Emx1 is amarker for pyramidal neurons of the cerebral cortex. CerebCortex11,1191-1198.


https://doi.org/10.1101/738443

32

Conrad,M.S.,Dilger,R.N.,andJohnson,R.W.(2012).Braingrowthofthedomesticpig (Susscrofa) from2to24weeksofage:a longitudinalMRIstudy.DevNeurosci34,291-298.Conrad,M.S.,Sutton,B.P.,Dilger,R.N.,andJohnson,R.W.(2014).Aninvivothree-dimensional magnetic resonance imaging-based averaged brain collection of theneonatalpiglet(Susscrofa).PLoSOne9,e107650.Deng, J.B., Yu,D.M., and Li,M.S. (2006). Formation of the entorhino-hippocampalpathway:atracingstudyinvitroandinvivo.NeurosciBull22,305-314.Diacou,R.,Zhao,Y.,Zheng,D.,Cvekl,A.,andLiu,W.(2018).Six3andSix6AreJointlyRequired for the Maintenance of Multipotent Retinal Progenitors through BothPositiveandNegativeRegulation.CellRep25,2510-2523e2514.Diekmann, S., Nitsch, R., and Ohm, T.G. (1994). The organotypic entorhinal-hippocampalcomplexslicecultureofadolescentrats.Amodeltostudytranscellularchangesinacircuitparticularlyvulnerableinneurodegenerativedisorders.JNeuralTransmSuppl44,61-71.Dobin, A., Davis, C.A., Schlesinger, F., Drenkow, J., Zaleski, C., Jha, S., Batut, P.,Chaisson,M., andGingeras, T.R. (2013). STAR: ultrafast universal RNA-seq aligner.Bioinformatics29,15-21.Domnisoru, C., Kinkhabwala, A.A., and Tank, D.W. (2013). Membrane potentialdynamicsofgridcells.Nature495,199-204.Donato, F., Jacobsen, R.I.,Moser,M.B., andMoser, E.I. (2017). Stellate cells drivematurationoftheentorhinal-hippocampalcircuit.Science355.Englund, C., Fink, A., Lau, C., Pham,D., Daza, R.A., Bulfone, A., Kowalczyk, T., andHevner,R.F. (2005).Pax6,Tbr2,andTbr1areexpressedsequentiallyby radialglia,intermediate progenitor cells, and postmitotic neurons in developing neocortex. JNeurosci25,247-251.Fan,X.,Dong,J.,Zhong,S.,Wei,Y.,Wu,Q.,Yan,L.,Yong,J.,Sun,L.,Wang,X.,Zhao,Y.,etal. (2018).Spatial transcriptomicsurveyofhumanembryoniccerebralcortexbysingle-cellRNA-seqanalysis.CellRes28,730-745.Fang,M.,Li,J.,Gong,X.,Antonio,G.,Lee,F.,Kwong,W.H.,Wai,S.M.,andYew,D.T.(2005). Myelination of the pig's brain: a correlated MRI and histological study.Neurosignals14,102-108.Farr, M., Kitas, G.D., Waterhouse, L., Jubb, R., Felix-Davies, D., and Bacon, P.A.(1988).Treatmentofpsoriaticarthritiswithsulphasalazine:aoneyearopenstudy.ClinRheumatol7,372-377.Fietz,S.A.,Kelava,I.,Vogt,J.,Wilsch-Brauninger,M.,Stenzel,D.,Fish,J.L.,Corbeil,D.,Riehn,A.,Distler,W.,Nitsch,R.,etal.(2010).OSVZprogenitorsofhumanandferretneocortexareepithelial-likeandexpandbyintegrinsignaling.NatNeurosci13,690-699.Fuchs,E.C.,Neitz,A.,Pinna,R.,Melzer,S.,Caputi,A.,andMonyer,H. (2016).LocalandDistant InputControllingExcitation inLayer IIof theMedialEntorhinalCortex.Neuron89,194-208.Fujimaru,Y.,andKosaka,T.(1996).Thedistributionoftwocalciumbindingproteins,calbindin D-28K and parvalbumin, in the entorhinal cortex of the adult mouse.NeurosciRes24,329-343.


https://doi.org/10.1101/738443

33

Fukai,N.,Eklund,L.,Marneros,A.G.,Oh,S.P.,Keene,D.R.,Tamarkin,L.,Niemela,M.,Ilves,M.,Li,E.,Pihlajaniemi,T.,etal.(2002).LackofcollagenXVIII/endostatinresultsineyeabnormalities.EMBOJ21,1535-1544.Garcia-Cabezas,M.A.,John,Y.J.,Barbas,H.,andZikopoulos,B.(2016).DistinctionofNeurons,Glia andEndothelialCells in theCerebralCortex:AnAlgorithmBasedonCytologicalFeatures.FrontNeuroanat10,107.Gatome,C.W.,Slomianka,L.,Lipp,H.P.,andAmrein,I.(2010).Numberestimatesofneuronal phenotypes in layer II of themedial entorhinal cortexof rat andmouse.Neuroscience170,156-165.Georgala, P.A., Manuel, M., and Price, D.J. (2011). The generation of superficialcorticallayersisregulatedbylevelsofthetranscriptionfactorPax6.CerebCortex21,81-94.Germroth,P.,Schwerdtfeger,W.K.,andBuhl,E.H.(1989).Morphologyof identifiedentorhinal neurons projecting to the hippocampus. A light microscopical studycombiningretrogradetracingandintracellularinjection.Neuroscience30,683-691.Giocomo,L.M.,Stensola,T.,Bonnevie,T.,VanCauter,T.,Moser,M.B.,andMoser,E.I.(2014).Topographyofheaddirectioncellsinmedialentorhinalcortex.CurrBiol24,252-262.Guo,F.,Maeda,Y.,Ma,J.,Xu,J.,Horiuchi,M.,Miers,L.,Vaccarino,F.,andPleasure,D.(2010).Pyramidalneuronsaregeneratedfromoligodendroglialprogenitorcellsinadultpiriformcortex.JNeurosci30,12036-12049.Haghverdi, L., Lun, A.T.L.,Morgan,M.D., andMarioni, J.C. (2018). Batch effects insingle-cell RNA-sequencing data are corrected by matching mutual nearestneighbors.NatBiotechnol36,421-427.Hering,T.M.,Wirthlin,L.,Ravindran,S.,andMcAlinden,A.(2014).ChangesintypeIIprocollagen isoform expression during chondrogenesis by disruption of analternative5'splicesitewithinCol2a1exon2.MatrixBiol36,51-63.Hevner,R.F.,Neogi,T.,Englund,C.,Daza,R.A.,andFink,A.(2003).Cajal-Retziuscellsin the mouse: transcription factors, neurotransmitters, and birthdays suggest apallialorigin.BrainResDevBrainRes141,39-53.Holm, I.E., and Geneser, F.A. (1989). Histochemical demonstration of zinc in thehippocampal region of the domestic pig: I. Entorhinal area, parasubiculum, andpresubiculum.JCompNeurol287,145-163.Hu,J.G.,Wang,Y.X.,Zhou,J.S.,Chen,C.J.,Wang,F.C.,Li,X.W.,andLu,H.Z. (2011).Differential gene expression in oligodendrocyte progenitor cells, oligodendrocytesandtypeIIastrocytes.TohokuJExpMed223,161-176.Insausti, R., Herrero, M.T., and Witter, M.P. (1997). Entorhinal cortex of the rat:cytoarchitectonic subdivisions and the origin and distribution of cortical efferents.Hippocampus7,146-183.Insausti,R.,Tunon,T., Sobreviela,T., Insausti,A.M.,andGonzalo, L.M. (1995).Thehumanentorhinalcortex:acytoarchitectonicanalysis.JCompNeurol355,171-198.Jelsing, J.,Nielsen, R.,Olsen,A.K.,Grand,N.,Hemmingsen,R., andPakkenberg, B.(2006). The postnatal development of neocortical neurons and glial cells in theGottingenminipigandthedomesticpigbrain.JExpBiol209,1454-1462.Karunaratne, A., Hargrave, M., Poh, A., and Yamada, T. (2002). GATA proteinsidentifyanovelventralinterneuronsubclassinthedevelopingchickspinalcord.DevBiol249,30-43.


https://doi.org/10.1101/738443

34

Kessaris,N.,Fogarty,M.,Iannarelli,P.,Grist,M.,Wegner,M.,andRichardson,W.D.(2006). Competing waves of oligodendrocytes in the forebrain and postnataleliminationofanembryoniclineage.NatNeurosci9,173-179.Kim,S.W.(2010).Recentadvances insownutrition.RevistaBrasileiradeZootecnia39,303-310.Kiselev,V.Y.,Yiu,A.,andHemberg,M.(2018).scmap:projectionofsingle-cellRNA-seqdataacrossdatasets.NatMethods15,359-362.Kitamura,T.,Pignatelli,M.,Suh,J.,Kohara,K.,Yoshiki,A.,Abe,K.,andTonegawa,S.(2014).Islandcellscontroltemporalassociationmemory.Science343,896-901.Kolenkiewicz, M., Robak, A., Rowniak, M., Bogus-Nowakowska, K., Calka, J., andMajewski,M.(2009).Distributionofcocaine-andamphetamine-regulatedtranscriptin the hippocampal formation of the guinea pig and domestic pig. FoliaMorphol(Warsz)68,23-31.Kongkham,P.N.,Northcott,P.A.,Croul,S.E.,Smith,C.A.,Taylor,M.D.,andRutka,J.T.(2010).TheSFRPfamilyofWNTinhibitorsfunctionasnoveltumorsuppressorgenesepigeneticallysilencedinmedulloblastoma.Oncogene29,3017-3024.Kostovic, I., Petanjek, Z., and Judas, M. (1993). Early areal differentiation of thehumancerebralcortex:entorhinalarea.Hippocampus3,447-458.Kostovic,I.,Sedmak,G.,andJudas,M.(2019).Neuralhistologyandneurogenesisofthehumanfetalandinfantbrain.Neuroimage188,743-773.Krishnaswami,S.R.,Grindberg,R.V.,Novotny,M.,Venepally,P.,Lacar,B.,Bhutani,K.,Linker, S.B., Pham, S., Erwin, J.A.,Miller, J.A., et al. (2016). Using single nuclei forRNA-seqtocapturethetranscriptomeofpostmortemneurons.NatProtoc11,499-524.Kropff, E., Carmichael, J.E.,Moser,M.B., andMoser, E.I. (2015). Speed cells in themedialentorhinalcortex.Nature523,419-424.Lagutin, O.V., Zhu, C.C., Kobayashi, D., Topczewski, J., Shimamura, K., Puelles, L.,Russell,H.R.,McKinnon,P.J.,Solnica-Krezel,L.,andOliver,G.(2003).Six3repressionofWntsignalingintheanteriorneuroectodermisessentialforvertebrateforebraindevelopment.GenesDev17,368-379.Langston,R.F.,Ainge,J.A.,Couey,J.J.,Canto,C.B.,Bjerknes,T.L.,Witter,M.P.,Moser,E.I., andMoser,M.B. (2010).Developmentof the spatial representation system intherat.Science328,1576-1580.Lein, E.S., Hawrylycz,M.J., Ao,N., Ayres,M., Bensinger, A., Bernard, A., Boe, A.F.,Boguski,M.S.,Brockway,K.S.,Byrnes,E.J.,etal.(2007).Genome-wideatlasofgeneexpressionintheadultmousebrain.Nature445,168-176.Leitner, F.C., Melzer, S., Lutcke, H., Pinna, R., Seeburg, P.H., Helmchen, F., andMonyer,H.(2016).Spatiallysegregatedfeedforwardandfeedbackneuronssupportdifferential odor processing in the lateral entorhinal cortex.NatNeurosci 19, 935-944.Lin,G.,Mela,A.,Guilfoyle,E.M.,andGoldman,J.E.(2009).NeonatalandadultO4(+)oligodendrocytelineagecellsdisplaydifferentgrowthfactorresponsesanddifferentgeneexpressionpatterns.JNeurosciRes87,3390-3402.Lyuksyutova,A.I., Lu,C.C.,Milanesio,N., King, L.A.,Guo,N.,Wang, Y.,Nathans, J.,Tessier-Lavigne,M.,andZou,Y.(2003).Anterior-posteriorguidanceofcommissuralaxonsbyWnt-frizzledsignaling.Science302,1984-1988.


https://doi.org/10.1101/738443

35

Magri,L.,Gacias,M.,Wu,M.,Swiss,V.A.,Janssen,W.G.,andCasaccia,P.(2014).c-Myc-dependent transcriptional regulation of cell cycle and nucleosomal histonesduringoligodendrocytedifferentiation.Neuroscience276,72-86.Martinez-Cerdeno, V., Cunningham, C.L., Camacho, J., Antczak, J.L., Prakash, A.N.,Cziep, M.E., Walker, A.I., and Noctor, S.C. (2012). Comparative analysis of thesubventricularzoneinrat,ferretandmacaque:evidenceforanoutersubventricularzoneinrodents.PLoSOne7,e30178.Miao, C., Cao, Q., Moser, M.B., and Moser, E.I. (2017). Parvalbumin andSomatostatin InterneuronsControlDifferent Space-CodingNetworks in theMedialEntorhinalCortex.Cell171,507-521e517.Mufson,E.J.,Brady,D.R.,andKordower,J.H.(1990).TracingneuronalconnectionsinpostmortemhumanhippocampalcomplexwiththecarbocyaninedyeDiI.NeurobiolAging11,649-653.Nilssen,E.S.,Jacobsen,B.,Fjeld,G.,Nair,R.R.,Blankvoort,S.,Kentros,C.,andWitter,M.P. (2018). Inhibitory ConnectivityDominates the FanCellNetwork in Layer II ofLateralEntorhinalCortex.JNeurosci38,9712-9727.Olton, D.S., Walker, J.A., and Wolf, W.A. (1982). A disconnection analysis ofhippocampalfunction.BrainRes233,241-253.Patten,A.R.,Fontaine,C.J.,andChristie,B.R. (2014).Acomparisonof thedifferentanimalmodelsoffetalalcoholspectrumdisordersandtheiruseinstudyingcomplexbehaviors.FrontPediatr2,93.Perez-Garcia, C.G., Gonzalez-Delgado, F.J., Suarez-Sola, M.L., Castro-Fuentes, R.,Martin-Trujillo,J.M.,Ferres-Torres,R.,andMeyer,G.(2001).Reelin-immunoreactiveneuronsintheadultvertebratepallium.JChemNeuroanat21,41-51.Pollen, A.A., Nowakowski, T.J., Chen, J., Retallack, H., Sandoval-Espinosa, C.,Nicholas, C.R., Shuga, J., Liu, S.J., Oldham,M.C., Diaz, A., et al. (2015).Molecularidentityofhumanouterradialgliaduringcorticaldevelopment.Cell163,55-67.Pontelo,T.P.,Miranda, J.R., Felix,M.A.R.,Pereira,B.A.,daSilva,W.E.,Avelar,G.F.,Mariano, F., Guimaraes, G.C., and Zangeronimo, M.G. (2018). Histologicalcharacteristics of the gonads of pig fetuses and their relationship with fetalanatomicalmeasurements.ResVetSci117,28-36.Quirk,G.J.,Muller,R.U.,Kubie, J.L.,andRanck, J.B., Jr. (1992).Thepositional firingproperties of medial entorhinal neurons: description and comparison withhippocampalplacecells.JNeurosci12,1945-1963.Ray,S.,andBrecht,M.(2016).Structuraldevelopmentanddorsoventralmaturationofthemedialentorhinalcortex.Elife5,e13343.Ridley, R.M., Samson, N.A., Baker, H.F., and Johnson, J.A. (1988). Visuospatiallearning impairment following lesion of the cholinergic projection to thehippocampus.BrainRes456,71-87.Rosenberg,A.B.,Roco,C.M.,Muscat,R.A.,Kuchina,A.,Sample,P.,Yao,Z.,Graybuck,L.T., Peeler,D.J.,Mukherjee, S., Chen,W., et al. (2018). Single-cell profiling of thedevelopingmousebrainandspinalcordwithsplit-poolbarcoding.Science360,176-182.Rowland,D.C.,Obenhaus,H.A., Skytoen,E.R., Zhang,Q.,Kentros,C.G.,Moser,E.I.,andMoser,M.B. (2018).Functionalpropertiesofstellatecells inmedialentorhinalcortexlayerII.Elife7.


https://doi.org/10.1101/738443

36

Rowland, D.C.,Weible, A.P.,Wickersham, I.R.,Wu, H.,Mayford,M.,Witter,M.P.,andKentros,C.G.(2013).Transgenicallytargetedrabiesvirusdemonstratesamajormonosynaptic projection fromhippocampal area CA2 tomedial entorhinal layer IIneurons.JNeurosci33,14889-14898.Roy,K.,Kuznicki,K.,Wu,Q.,Sun,Z.,Bock,D.,Schutz,G.,Vranich,N.,andMonaghan,A.P. (2004). The Tlx gene regulates the timing of neurogenesis in the cortex. JNeurosci24,8333-8345.Saikali,S.,Meurice,P.,Sauleau,P.,Eliat,P.A.,Bellaud,P.,Randuineau,G.,Verin,M.,andMalbert, C.H. (2010). A three-dimensional digital segmented and deformablebrainatlasofthedomesticpig.JNeurosciMethods192,102-109.Sano, N., Shimogawa, T., Sakaguchi, H., Ioroi, Y., Miyawaki, Y., Morizane, A.,Miyamoto, S., and Takahashi, J. (2017). Enhanced Axonal Extension of SubcorticalProjectionNeuronsIsolatedfromMurineEmbryonicCortexusingNeuropilin-1.FrontCellNeurosci11,123.Sansom, S.N., Griffiths, D.S., Faedo, A., Kleinjan, D.J., Ruan, Y., Smith, J., vanHeyningen,V.,Rubenstein,J.L.,andLivesey,F.J.(2009).Thelevelofthetranscriptionfactor Pax6 is essential for controlling the balance between neural stem cell self-renewalandneurogenesis.PLoSGenet5,e1000511.Sargolini,F.,Fyhn,M.,Hafting,T.,McNaughton,B.L.,Witter,M.P.,Moser,M.B.,andMoser,E.I.(2006).Conjunctiverepresentationofposition,direction,andvelocityinentorhinalcortex.Science312,758-762.Satija, R., Farrell, J.A., Gennert, D., Schier, A.F., and Regev, A. (2015). Spatialreconstructionofsingle-cellgeneexpressiondata.NatBiotechnol33,495-502.Schmidt-Hieber, C., and Hausser, M. (2013). Cellular mechanisms of spatialnavigationinthemedialentorhinalcortex.NatNeurosci16,325-331.Seress, L., Leranth, C., and Frotscher, M. (1994). Distribution of calbindin D28kimmunoreactivecellsandfibersinthemonkeyhippocampus,subicularcomplexandentorhinalcortex.Alightandelectronmicroscopicstudy.JHirnforsch35,473-486.Smart, I.H., Dehay, C., Giroud, P., Berland, M., and Kennedy, H. (2002). Uniquemorphological featuresoftheproliferativezonesandpostmitoticcompartmentsofthe neural epithelium giving rise to striate and extrastriate cortex in themonkey.CerebCortex12,37-53.Solstad, T., Boccara, C.N., Kropff, E., Moser, M.B., and Moser, E.I. (2008).Representation of geometric borders in the entorhinal cortex. Science 322, 1865-1868.Stagni, F., Giacomini, A., Guidi, S., Ciani, E., and Bartesaghi, R. (2015). Timing oftherapiesforDownsyndrome:thesooner,thebetter.FrontBehavNeurosci9,265.Stensola,T.,andMoser,E.I.(2016).GridCellsandSpatialMapsinEntorhinalCortexandHippocampus. InMicro-,Meso- andMacro-Dynamicsof theBrain,G.Buzsaki,andY.Christen,eds.(Cham(CH)),pp.59-80.Sun,L.,Wang,J.,Liu,H.,Fan,Z.,Wang,S.,andDu,J.(2017).AComprehensiveStudyofPalateDevelopmentinMiniaturePig.AnatRec(Hoboken)300,1409-1419.Sun,Y.,Nguyen,A.Q.,Nguyen,J.P.,Le,L.,Saur,D.,Choi,J.,Callaway,E.M.,andXu,X.(2014). Cell-type-specific circuit connectivityofhippocampalCA1 revealed throughCre-dependentrabiestracing.CellRep7,269-280.


https://doi.org/10.1101/738443

37

Tang,Q.,Burgalossi,A.,Ebbesen,C.L.,Ray,S.,Naumann,R.,Schmidt,H.,Spicher,D.,and Brecht, M. (2014). Pyramidal and stellate cell specificity of grid and borderrepresentationsinlayer2ofmedialentorhinalcortex.Neuron84,1191-1197.Taube, J.S.,Muller, R.U., and Ranck, J.B., Jr. (1990). Head-direction cells recordedfrom the postsubiculum in freely moving rats. II. Effects of environmentalmanipulations.JNeurosci10,436-447.Tirosh,I.,Izar,B.,Prakadan,S.M.,Wadsworth,M.H.,2nd,Treacy,D.,Trombetta,J.J.,Rotem,A.,Rodman,C.,Lian,C.,Murphy,G.,etal.(2016).Dissectingthemulticellularecosystemofmetastaticmelanomabysingle-cellRNA-seq.Science352,189-196.Totterdell, S.,andMeredith,G.E. (1997).Topographicalorganizationofprojectionsfromtheentorhinalcortextothestriatumoftherat.Neuroscience78,715-729.vanderMaaten, L., andHinton,G. (2008). Visualizingdatausing t-SNE. Journal ofMachineLearningResearch9,2579-2605.Varga, C., Lee, S.Y., and Soltesz, I. (2010). Target-selective GABAergic control ofentorhinalcortexoutput.NatNeurosci13,822-824.Wang, Y., Dye, C.A., Sohal, V., Long, J.E., Estrada, R.C., Roztocil, T., Lufkin, T.,Deisseroth,K.,Baraban,S.C.,andRubenstein,J.L.(2010).Dlx5andDlx6regulatethedevelopmentofparvalbumin-expressingcortical interneurons. JNeurosci30,5334-5345.Wilhite,B.L.,Teyler,T.J.,andHendricks,C.(1986).Functionalrelationsoftherodentclaustral-entorhinal-hippocampalsystem.BrainRes365,54-60.Willems, J.G.P., Wadman, W.J., and Cappaert, N.L.M. (2018). Parvalbumininterneuron mediated feedforward inhibition controls signal output in the deeplayersoftheperirhinal-entorhinalcortex.Hippocampus28,281-296.Winter, J.D., Dorner, S., Lukovic, J., Fisher, J.A., St Lawrence, K.S., and Kassner, A.(2011).NoninvasiveMRImeasuresofmicrostructural andcerebrovascular changesduringnormalswinebraindevelopment.PediatrRes69,418-424.Winterer, J.,Maier,N.,Wozny,C.,Beed,P.,Breustedt, J., Evangelista,R.,Peng,Y.,D'Albis, T., Kempter, R., and Schmitz, D. (2017). Excitatory Microcircuits withinSuperficialLayersoftheMedialEntorhinalCortex.CellRep19,1110-1116.Witter, M.P., Doan, T.P., Jacobsen, B., Nilssen, E.S., and Ohara, S. (2017).Architecture of the Entorhinal Cortex A Review of Entorhinal Anatomy in RodentswithSomeComparativeNotes.FrontSystNeurosci11,46.Witter,M.P., and Groenewegen, H.J. (1986). Connections of the parahippocampalcortexinthecat.IV.Subcorticalefferents.JCompNeurol252,51-77.Witter,M.P.,Room,P.,Groenewegen,H.J.,andLohman,A.H.(1986).Connectionsofthe parahippocampal cortex in the cat. V. Intrinsic connections; comments oninput/outputconnectionswiththehippocampus.JCompNeurol252,78-94.Wouterlood, F.G., Hartig,W., Bruckner, G., andWitter,M.P. (1995). Parvalbumin-immunoreactive neurons in the entorhinal cortex of the rat: localization,morphology,connectivityandultrastructure.JNeurocytol24,135-153.Woznicka,A.,Malinowska,M.,andKosmal,A.(2006).Cytoarchitectonicorganizationoftheentorhinalcortexofthecaninebrain.BrainResRev52,346-367.Wyss, J.M., Sripanidkulchai, B., and Hickey, T.L. (1983). An analysis of the time oforiginofneuronsintheentorhinalandsubicularcorticesofthecat.JCompNeurol221,341-357.


https://doi.org/10.1101/738443

38

Ye, J., Witter, M.P., Moser, M.B., and Moser, E.I. (2018). Entorhinal fast-spikingspeedcellsprojecttothehippocampus.ProcNatlAcadSciUSA115,E1627-E1636.Yushkevich,P.A.,Piven,J.,Hazlett,H.C.,Smith,R.G.,Ho,S.,Gee,J.C.,andGerig,G.(2006). User-guided 3D active contour segmentation of anatomical structures:significantlyimprovedefficiencyandreliability.Neuroimage31,1116-1128.Zhong,S.,Zhang,S.,Fan,X.,Wu,Q.,Yan,L.,Dong,J.,Zhang,H.,Li,L.,Sun,L.,Pan,N.,et al. (2018). A single-cell RNA-seq survey of the developmental landscape of thehumanprefrontalcortex.Nature555,524-528.


https://doi.org/10.1101/738443

E100M

EC

LEC

PE

RN

EO

E

10 μm

D

100 μm

E80 E100 P75

ME

CLE

CP

ER

NE

O

LIVBorderM L

C

MEC

MEC

LEC

LEC

MEC

PaS

LEC

PER

L2L3

L5

L2L3

L5

L2

L3

L5

L2L3

L5

PreS

500 µm

1 cm

Olf.Tr.COA

RHP

S.Sag

HA

MEC

LEC

B

0

0

0

10.5 20

29498 196

11450 70

1 trimester 2 trimester 3 trimesterst nd rd

17

A


https://doi.org/10.1101/738443

A 1 2 34 56

1 2 3 4 5 6

1 2 3 4 5 6

1 2 3 4 5 6

E60

P75

P

D

B

EC

: C

orte

x vo

lum

e

Cortex volume EC to Cortex ratioEC volume

Y = 0.005871X - 0.3192

R2 = 0.9996

cm3

0

20

40

60

0.0

0.2

0.4

0.6

0.8

0.00

0.01

0.02

0.0

0.2

0.4

0.6

0.8** **

Days post fertilization

EC volume change

cm3

E60

E70

E80

E10

0

P75

E60

E70

E80

E10

0

P75

E60

E70

E80

E10

0

P75 60 80 10

012

014

016

018

0

cm3

C

A

D

Connection of EC white matter fibers

MEC LEC

D

MEC LEC0

2000

4000

6000

8000

10000

Num

ber o

f whi

te m

atte

r fib

ers

Total emerging neuronal tractsn.s.

E

A

D

Peri- and post-rhinal cortex

A

M

F Corpus collosum

M

D

G Olfactory bulb

H Hippocampus

A

M

I Amygdala

M

D

CA3DG PUT

LOT

CC EC AM CA1SUB

050100150200250500

1000

1500MECLEC

J

Whi

te m

atte

r fib

ers

Neuronal tract connectivity

n.s.n.s.n.s. n.s.n.s.n.s.

n.s.

n.s.

n.s.


https://doi.org/10.1101/738443

E60-MEC E60-DE50-V E50-D

SA

TB2

/B

CL1

1B/

RE

LNA C

D

V

E50 E60

B

MEC

MEC

1

2

3

45

LEC

Rhinal cortex

1

2

3

4

5

E70-EC E80-EC P75-ECE100-EC

D

E50 E60

D

Late born BCL11B neuron Early born BCL11B neuron SATB2 neuron

Coronal view E

SA

TB2

/B

CL1

1B/

RE

LN

E70-

MEC

E80-

MEC

E100

-MEC

P75-

MEC

M

D

M

D

50 μm 50 μm

25 μm


https://doi.org/10.1101/738443

Whole cells

Isolated nucleusEC

Single-cellRNA-seq

000000001094740274010001020000435070657023404505054602450200002340001002040606000020204060600303104035600012304530570012023000030140425034000000020030050023040230120000010200301030204003010100304003020102040505030010010030040500301010002030404050600101102004050606070600202010303040506050002011

000000001094740274010001020000435070657023404505054602450200002340001002040606000020204060600303104035600012304530570012023000030140425034000000020030050023040230120000010200301030204003010100304003020102040505030010010030040500301010002030404050600101102004050606070600202010303040506050002011

000000001094740274010001020000435070657023404505054602450200002340001002040606000020204060600303104035600012304530570012023000030140425034000000020030050023040230120000010200301030204003010100304003020102040505030010010030040500301010002030404050600101102004050606070600202010303040506050002011

000000001094740274010001020000435070657023404505054602450200002340001002040606000020204060600303104035600012304530570012023000030140425034000000020030050023040230120000010200301030204003010100304003020102040505030010010030040500301010002030404050600101102004050606070600202010303040506050002011

000000001094740274010001020000435070657023404505054602450200002340001002040606000020204060600303104035600012304530570012023000030140425034000000020030050023040230120000010200301030204003010100304003020102040505030010010030040500301010002030404050600101102004050606070600202010303040506050002011

Merge and batch correctFast MNN

Dimensional reduction& clustering in Seurat

a b c d e f g h j

a b c d e f g h j

1234

1234

Subclustering& analysis

A n = 8

n = 2

B

t-SNE 1

t-SN

E 2

n cells = 24,294

17

0

5

10

15 1921 22

23

2026

28

8

911

129

134

6

7

14

31

3

24

27 12

30

2518

162

10 20 300

n UMIx1000

50249145

1,677822

1,023862105

1,5451,753

645727

1,317272292275425

2,9241,431

91653060770

1,677248448568239398120926

1,008

n cells

0 2 4 6

n genesx1000

62321221905

101715302

25181626202898

3124273

127

112941

1413

Olig

oG

lial c

ells

IPE

xcita

tory

INM

icro

glia

MB

PC

LDN

11O

LIG

1P

DG

FRA

AQ

P4

GFA

PG

LAS

TH

ES

1S

OX

2PA

X6

EO

ME

S

TBR

1E

MX

1S

ATB

2C

TIP

2V

GLU

T2V

GLU

T1D

CX

DLX

1G

AD

1G

AD

2A

IF1

CX

CR

3P

TPR

CC

SF1

R

NE

UR

OD

1C

202-Expression

t-SNE 1

t-SN

E 2

D

Who

le-c

ell

Nuc

lei

E50_1 E60_1 E60_2 E60_3

E70_1 E70_2 E70_3 Adult_1

E70_nc

Adult_nc

Batch

t-SNE 1

t-SN

E 2

EOligo OPC Astrocytes NPC IN

Microglia HBB UnassignedExcitatory

scmap projections


https://doi.org/10.1101/738443

OLI3

OPC0OPC1

OLI2

OLI4

t-SNE 1

t-SN

E 2

n cells = 2,696

A B

t-SNE 1t-S

NE

2n cells = 2,696

E50

E60

E70

Adult

Lineage markers OPC0 OPC1 OLI2 OLI3 OLI4

OPC0

OPC1

OLI2

OLI3

OLI4

PD

GFR

AO

LIG

1O

LIG

2FA

BP

7C

LDN

11M

OG

GP

R17

NK

X2-

2M

AG

TF *UN

PR

OT2

DLG

2*M

OB

PS

LC44

A1

MM

P16

LHFP

L3C

NTN

1D

BI

*UN

PR

OT1

OP

ALI

N*G

LUL

*AN

K3

RA

SS

F8O

XR1

TUB

BS

TMN

1P

TMA

CA

LM2

*HM

GB

1B

CA

S1

DY

NLL

1FY

NFR

MD

4AM

BP

1- 0 1 2Normalized expression % cells expressing gene in cluster 0 50 7525 100C1.00

0.75

0.50

0.25

0.00

Freq

uenc

y

S-p

hase

G2M

-pha

seG

1-ph

ase

626

194

162

515

1199 n cells

OP

C0

OP

C1

OLI

2

OLI

3

OLI

4

D

1000 µm

500 µm

CNTN1 FYN

OXR1 DLG2

E

E18

.5A

dult


https://doi.org/10.1101/738443

NE

UR

OD

6TN

RS

FRP

2

EP

HA

6

1000 µm

RE

LNC

PN

E4

D Pyramidal cells

Speed cells

Reelin cells

500 µmCX

CR

4

1000 µm

1000 µm

Adu

ltA

dult

Adu

lt

Adu

ltA

dult

Adu

ltE

18.5

SPE9 RELN8 PYR7 PYR6 RELN5 IP4 PYR3 PYR2 PYR12 RELN11 RELN10 PYR1 PYR0Mark

0 50 7525 100

SFR

P2C

OL2

A1C

OL1

8A1

SIX3

SIX6

ZIC

2*P

ILR

BZI

C1

CBL

N1

ZIC

4H

PCAL

4D

IRAS

2R

BP4

LRR

TM1

ARH

GAP

20R

SPO

3AR

HG

DIB

NTS

GPR

22TC

F7L2

ND

ST4

TFAP

2DPO

U4F

2LH

X9R

ND

3D

MR

TA2

SOX2

EOM

ESN

EUR

OD

4N

PY1R

CAM

K2B

NTF

3N

UD

T14

ARPP

21EL

MO

D1

FAM

184B

PID

1TN

RN

EUR

OD

6*T

TC37

SALL

1D

LX1

ARX

DLX

5H

SD11

B2C

DC

20PB

KTI

MP1

CXC

R4

PPW

D1

CD

80*U

NPR

OT3

*USP

6NL

GTS

F1C

Xorf5

7G

LI3

CC

DC

173

RFX

2FA

BP7

KDSR

SEM

A3C

DPY

19L1

NR

P1

REL

NvG

LUT2

EMX1

TBR

1

EPH

A6

*CPN

E4PYR0PYR1PYR2PYR3IP4RELN5PYR6PYR7RELN8SPE9RELN10RELN11PYR12

C Normalized expression % cells expressing gene in cluster1- 0 1 2

B

0.00

0.25

0.50

0.75

1.00

Freq

uenc

y

PY

R0

PY

R1

PY

R2

PY

R3

IP4

RE

LN5

PY

R6

PY

R7

RE

LN8

SP

E9

RE

LN10

RE

LN11

PY

R12

729 641515

17112632

469

169

251

428

46

213

471

78

590

306

390

105

268

172

193

161

130

118

151

231

29

199

70

37

66

Adu

ltE

70E

60E

50

1435

1312

1209 64

557

751

639

137

229

127

427

326

910

8 n cellsA

t-SNE 1

t-SN

E 2

n cells = 7,672

RELN11

RELN8

RELN5

IP4

PYR12PYR1

PYR0

RELN10

PYR2

PYR3

PYR7

PYR6

SPE9


https://doi.org/10.1101/738443

C

0.00

0.25

0.50

0.75

1.00

Freq

uenc

y

IN0

IN1

IN2

IN3

IN4

IN5

IN6

KIT

LGN

GFR

*UN

PR

OT4

STK

17B

FOX

P2

ND

NF

ITP

R1

AR

HG

EF3

3C

A8

CA

LB1

RFX

2R

RO

X1

*TM

SB

4XTH

RB

ZBTB

16N

R2F

2D

LX5

PD

ZRN

3TO

X3

CH

D7

GA

TA3

TEX

14E

MX

2N

FIA

MA

ML2

NFI

BN

FIX

ZFP

M2

DLX

1V

WC

2C

UX

2A

RX

NX

PH

2S

OX

6LH

X6

CO

RT

CX

CR

4M

AF

*UN

PR

OT5

KIF

26B

CD

H8

*KC

TD8

THS

D4

*UN

PR

OT6

*PC

LOG

RM

7C

NA

1DM

DG

A2

EN

HO

TME

M17

6BP

CB

D1

ISG

12(A

)S

PO

CK

2R

AB

6BD

US

P26

*ZD

HH

CL2

*RES

P18

CH

EA

PR

XL2

B*U

NP

RO

T7M

AP

1LC

3AG

PX

1LI

N7B

ALD

OC

GA

P43

GN

G3

MT3

SNC

B

D

SST / BCL11B / HO F

A

E

B

L1L2L3/4

L5/6

IN0

IN1

IN2

IN3

IN4

IN5

IN6

IN0

IN1

IN2

IN4

IN5

IN6

IN3

t-SNE 1

t-SN

E 2

n cells = 2,942

Adu

ltE

70E

60E

50

502

128

64

222

233 214

84

121

42

120

256

186

82

99

100

175

135

104

718

465

461

394 n cells37

728

723

9

LHX1

LHX6

NPY

RELN

SST

TACR1

CALB1

CALB2

DLX5

GAD1

GAD2

GATA3

IN0 IN1 IN2 IN3 IN4 IN5 IN6 IN0 IN1 IN2 IN3 IN4 IN5 IN6

0 50 7525 100Normalized expression % cells expressing gene in cluster1- 0 1 2

IN6 IN3 IN2 IN1 IN0IN5 IN4

AdultAdultAdult

Adult Adult E18.5

Sncb Kctd8 Ndnf

Resp18 Cort Ndnf

25 µm50 µm

200 µm

200 µm


https://doi.org/10.1101/738443

Sandwich cortical lamination and single-cell analysis ... · 8/20/2019 · 1 Title: Sandwich...

Documents

Transcript of Sandwich cortical lamination and single-cell analysis ... · 8/20/2019 · 1 Title: Sandwich...