The function of the heart as a pump is vital for the circu- lation of nutrients and the removal of metabolic waste as soon as the embryo reaches a size at which passive diffu- sion is no longer efficient. In the early embryo, the beating heart tube consists of the striated muscle of the myocar- dium, with an endothelial layer of cells, the endocardium, that lines the lumen (FIG. 1). The cardiac tube is initially attached by the dorsal mesocardium, situated under the head folds. The dorsal mesocardium then breaks down, leaving the cardiac tube suspended with attachment points to the pericardial wall, anteriorly at the arterial pole, where blood is pumped to the embryo, and posteri- orly at the venous pole, where blood enters the heart. The early cardiac tube grows by division of existing cardio- myocytes and by addition of myocardial progenitors, ini- tially dorsally and then to the poles of the heart 1–3 . As the cardiac tube elongates, it undergoes rightward looping 4 and convergence of the poles, which become juxtaposed, with the arterial pole lying ventrally. The cardiac chambers grow out from the looping tube 5 in a process described as ballooning 6 , leading to the formation of the right and left ventricles and the atria. As the chambers form, an epithe- lial layer of epicardium grows over the outer surface of the heart. The architecture of the cardiac muscle matures, with the formation of ridges, called trabeculae, that pro- ject into the lumen of the chambers, thus increasing the myocardial surface area to facilitate the uptake of oxygen and nutrients. This extended surface is no longer required when the coronary vascular system develops, and trabecu- lae subsequently undergo compaction. The myocardium is also invaded by interstitial cells such as fibroblasts, which deposit and maintain the extracellular matrix. Septation takes place to separate the cardiac chambers and to divide the outflow region and thus establish a double blood cir- culation. Valves develop to impose unidirectional blood flow between the different compartments. Coordination of muscle contraction is ensured by the formation of the cardiac conduction system. This complex process of cardiogenesis (FIG. 1), which takes place while the devel- oping heart is performing a vital function, requires very precise spatial and temporal coordination of cardiac cell populations. As many as 0.8% of children in developed countries are born with a congenital heart malforma- tion, which can be fatal if not corrected by cardiac sur- geons 7 . In this Review, we discuss the origin of cardiac cell populations, their lineage relationships and the genes that regulate their behaviour and differentiation. We focus on the mouse as a mammalian model that allows genetic approaches to dissect the different facets of cardiogenesis. Myocardial cell lineages Classic experiments with avian embryos have provided much of the initial background knowledge of the ori- gins of cells that contribute to cardiogenesis 8–14 . In the The deployment of cell lineages that form the mammalian heart Sigolène M. Meilhac 1,2 * and Margaret E. Buckingham 3 * Abstract | The function of the mammalian heart depends on the interplay between different cardiac cell types. The deployment of these cells, with precise spatiotemporal regulation, is also important during development to establish the heart structure. In this Review, we discuss the diverse origins of cardiac cell types and the lineage relationships between cells of a given type that contribute to different parts of the heart. The emerging lineage tree shows the progression of cell fate diversification, with patterning cues preceding cell type segregation, as well as points of convergence, with overlapping lineages contributing to a given tissue. Several cell lineage markers have been identified. However, caution is required with genetic-tracing experiments in comparison with clonal analyses. Genetic studies on cell populations provided insights into the mechanisms for lineage decisions. In the past 3 years, results of single-cell transcriptomics are beginning to reveal cell heterogeneity and early developmental trajectories. Equating this information with the in vivo location of cells and their lineage history is a current challenge. Characterization of the progenitor cells that form the heart and of the gene regulatory networks that control their deployment is of major importance for understanding the origin of congenital heart malformations and for producing cardiac tissue for use in regenerative medicine. 1 Imagine Institut Pasteur, Laboratory of Heart Morphogenesis, Paris, France. 2 INSERM UMR1163, Université Paris Descartes, Paris, France. 3 Department of Developmental and Stem Cell Biology, Centre National de la Recherche Scientifique (CNRS) UMR 3738, Institut Pasteur, Paris, France. *e-mail: sigolene.meilhac@ pasteur.fr; margaret.buckingham@ pasteur.fr https://doi.org/10.1038/ s41569-018-0086-9 REVIEWS NATURE REVIEWS | CARDIOLOGY
Welcome message from author
This document is posted to help you gain knowledge. Please leave a comment to let me know what you think about it! Share it to your friends and learn new things together.
Transcript
The function of the heart as a pump is vital for the circu-lation of nutrients and the removal of metabolic waste as soon as the embryo reaches a size at which passive diffu-sion is no longer efficient. In the early embryo, the beating heart tube consists of the striated muscle of the myocar-dium, with an endothelial layer of cells, the endocardium, that lines the lumen (Fig. 1). The cardiac tube is initially attached by the dorsal mesocardium, situated under the head folds. The dorsal mesocardium then breaks down, leaving the cardiac tube suspended with attachment points to the pericardial wall, anteriorly at the arterial pole, where blood is pumped to the embryo, and posteri-orly at the venous pole, where blood enters the heart. The early cardiac tube grows by division of existing cardio-myocytes and by addition of myocardial progenitors, ini-tially dorsally and then to the poles of the heart1–3. As the cardiac tube elongates, it undergoes rightward looping4 and convergence of the poles, which become juxtaposed, with the arterial pole lying ventrally. The cardiac chambers grow out from the looping tube5 in a process described as ballooning6, leading to the formation of the right and left ventricles and the atria. As the chambers form, an epithe-lial layer of epicardium grows over the outer surface of the heart. The architecture of the cardiac muscle matures, with the formation of ridges, called trabeculae, that pro-ject into the lumen of the chambers, thus increasing the myocardial surface area to facilitate the uptake of oxygen
and nutrients. This extended surface is no longer required when the coronary vascular system develops, and trabecu-lae subsequently undergo compaction. The myocardium is also invaded by interstitial cells such as fibroblasts, which deposit and maintain the extracellular matrix. Septation takes place to separate the cardiac chambers and to divide the outflow region and thus establish a double blood cir-culation. Valves develop to impose unidirectional blood flow between the different compartments. Coordination of muscle contraction is ensured by the formation of the cardiac conduction system. This complex process of cardiogenesis (Fig. 1), which takes place while the devel-oping heart is performing a vital function, requires very precise spatial and temporal coordination of cardiac cell populations. As many as 0.8% of children in developed countries are born with a congenital heart malforma-tion, which can be fatal if not corrected by cardiac sur-geons7. In this Review, we discuss the origin of cardiac cell populations, their lineage relationships and the genes that regulate their behaviour and differentiation. We focus on the mouse as a mammalian model that allows genetic approaches to dissect the different facets of cardiogenesis.
Myocardial cell lineagesClassic experiments with avian embryos have provided much of the initial background knowledge of the ori-gins of cells that contribute to cardiogenesis8–14. In the
The deployment of cell lineages that form the mammalian heartSigolène M. Meilhac1,2* and Margaret E. Buckingham3*
Abstract | The function of the mammalian heart depends on the interplay between different cardiac cell types. The deployment of these cells, with precise spatiotemporal regulation, is also important during development to establish the heart structure. In this Review , we discuss the diverse origins of cardiac cell types and the lineage relationships between cells of a given type that contribute to different parts of the heart. The emerging lineage tree shows the progression of cell fate diversification, with patterning cues preceding cell type segregation, as well as points of convergence, with overlapping lineages contributing to a given tissue. Several cell lineage markers have been identified. However, caution is required with genetic- tracing experiments in comparison with clonal analyses. Genetic studies on cell populations provided insights into the mechanisms for lineage decisions. In the past 3 years, results of single- cell transcriptomics are beginning to reveal cell heterogeneity and early developmental trajectories. Equating this information with the in vivo location of cells and their lineage history is a current challenge. Characterization of the progenitor cells that form the heart and of the gene regulatory networks that control their deployment is of major importance for understanding the origin of congenital heart malformations and for producing cardiac tissue for use in regenerative medicine.
1Imagine Institut Pasteur, Laboratory of Heart Morphogenesis, Paris, France.2INSERM UMR1163, Université Paris Descartes, Paris, France.3Department of Developmental and Stem Cell Biology, Centre National de la Recherche Scientifique (CNRS) UMR 3738, Institut Pasteur, Paris, France.
*e- mail: [email protected]; margaret.buckingham@ pasteur.fr
https://doi.org/10.1038/ s41569-018-0086-9
REVIEwS
Nature reviews | Cardiology
mouse embryo, the location of cardiac progenitor cells before gastrulation (Box 1) was mapped by label injec-tion into individual cells in the epiblast (at embryonic day (E) 6.5), followed by embryo culture to the early heart tube stage15,16. These cardiac progenitors in the epiblast are not yet committed to a cardiac fate, as they are clonally related to the endoderm, the neurecto-derm, the extra- embryonic mesoderm and the paraxial mesoderm15–17. Fate mapping showed that the first car-diac progenitors ingress early, at the mid- streak stage, and locate in the anterior region of the primitive streak, in close proximity to progenitors of the cranial meso-derm18. After gastrulation, cardiac progenitors within the lateral plate mesoderm (Box 1) move anteriorly to locate on either side of the embryo midline under the head folds (E7.5), where the first differentiated myo-cardial cells appear in an epithelial structure known as the cardiac crescent19,20. The cardiac crescent goes on to fuse at the midline to form the early heart tube. Grafting experiments have indicated that progenitor cells are committed to a cardiac fate when they have completed their migration in the cranial region of the E7.5 embryo21.
Clonal analyses reveal two cell lineages. The myo-cardium, which underlies cardiac contractile function, is also essential for shaping the structure of the heart. The first lineage study of myocardial progenitors in the mouse embryo was performed by retrospective clonal analysis (Box 2) with Actc1 (encoding actin, α- cardiac muscle 1) expression as the readout for rare cells that had activated an nlaacZ reporter22. This clonal analy-sis revealed that two distinct myocardial cell lineages contribute to the embryonic heart at E8.5; the first cell lineage forms the future left ventricle and contributes to other compartments of the heart with the exception of the outflow tract region, whereas the second cell lineage forms the myocardium of the outflow tract, the future right ventricle and part of the inflow myocardium that will form the atria (Fig. 2). On the basis of the number of β- galactosidase-positive cells in large clones, the lineage segregation was estimated to occur at or before the onset of gastrulation.
Genetic tracing (Box 2) is based on the use of a con-ditional reporter, the expression of which is activated by a system under the control of regulatory sequences of a gene marking the cell population under analysis. Mesp1, which encodes the basic helix–loop–helix transcription factor MESP1, is active in the primitive streak and in the nascent cardiac mesoderm. Mesp1–Cre genetic trac-ing leads to labelling of most (70%) cells in the heart23 as well as some other mesodermal derivatives, such as head muscles24. Clonal analysis of the cells expressing the marker of interest is possible with the use of an inducible system to control reporter gene expression25 (Box 2). For example, with a transgenic line containing the reverse- tetracycline transactivator (rtTA) under the control of Mesp1 regulatory sequences, crossed to a tetO–Cre;RosaConfetti reporter line, multicolour clonal analysis (Box 2) was performed with low levels of dox-ycycline induction at different time points26. At E12.5, cell labelling in the myocardium of the left ventricle was never observed together with cell labelling in the poles of the heart, a finding consistent with the exclusive contri-butions of the two myocardial cell lineages identified by retrospective clonal analysis. Early induction with doxy-cycline at E6.25–E6.75 led to cell labelling indicative of first lineage derivatives, whereas most second lineage derivatives were labelled after induction at E6.75–E7.25. Therefore, these results showed that first and second lin-eage progenitors can be temporarily distinguished and ingress sequentially through the Mesp1+ region of the primitive streak. Analyses of clonal dynamics led to the estimation that the myocardium originates from 250 Mesp1-expressing progenitors27.
In another study of cardiovascular progenitors, Mesp1–Cre was used with the mosaic analysis with double markers (MADM; also known as twin- spot) reporter, which generates very few labelled cells at clonal density28 (Box 2). Analysis of twin- spot derivatives in mosaic mice at E11.5–E14.5 showed no overlapping contribution of unilabelled cells to the myocardium of the left ventricle and outflow region of the heart, con-firming the contributions of two distinct myocardial cell lineages29. However, no clone was detected that spanned the left and right ventricles26,29, which was in contrast to previous clonal analyses at E8.5 that were not based on Mesp1 regulatory sequences22. This observation sug-gests that clones spanning the left and right ventricles do not derive from Mesp1-expressing progenitors or that cell sorting occurs, with separation of labelled clusters at the left–right boundary later during cardiac mor-phogenesis30. Further experiments identified enhancer sequences that regulate Smarcd3, which encodes a com-ponent of the BAF chromatin remodelling complex and is expressed in the cardiac mesoderm in a subset of cells that have expressed Mesp1 in the primitive streak29. The enhancer F1 was used to drive an inducible Smarcd3–F1–CreERT2 transgene in embryos with a conditional reporter. Using low doses of tamoxifen, isolated clusters of labelled cells in the heart showed a location consistent with first or second lineage contributions. As with the Mesp1-based clonal analyses, no evidence of a common progenitor for both lineages was obtained, consistent with the suggestion that such common progenitors,
Key points
•Clonalanalysisshowsthattwomyocardialcelllineages,whichsegregateearlyatgastrulation,formtheheart,withcellsublineagescontributingtothearterialandvenouspoles,withearlyleft–rightdelineation.
•Theoriginandcellfatechoicesofnon-myocardialprogenitors,suchasthosethatgiverisetocardiacinterstitialfibroblastsorthecoronaryvasculature,arenowclearer.
•Overlappingcellsourcesprovidepotentialforcompensatorymechanisms,andthusdevelopmentalrobustness,aprocessthatisjustbeginningtobecharacterized.
•Astheheartbeginstoform,cardiacprogenitorsarelocatedinthefirstandsecondheartfields,withcharacteristicanddiversegeneexpressionpatternsmarkingtheircardiaccontributions,whichcorrespondtothefirstandsecondmyocardiallineages.
•Generegulatorynetworks,governedbytranscriptionfactors,cofactorsandchromatinmodifications,inwhichnon-codingRNAsalsoparticipate,controlthedeploymentofcardiacprogenitorcellsduringcardiogenesis.
•Single-cellanalyseshaveidentifiedearlycardiacprogenitorcelltypes,providingnewinsightsintocellheterogeneityanddevelopmentaltrajectoriesofcardiaccellsastheheartdevelops.
www.nature.com/nrcardio
R e v i e w s
R e v i e w s
detected by retrospective clonal analysis, are present before activation of Mesp1 or Smarcd3, thus before gastrulation22,26,29.
Live imaging of early heart formation. Live- imaging experiments of mouse embryos at the time of heart tube formation have provided insight into the dynam-ics of myocardial differentiation19. Myocardial differ-entiation was shown to occur in two phases. The first wave occurs at the early head- fold stage during 5–6 h and corresponds to the formation of the cardiac cres-cent. A pause of 5–7 h is observed before myocardial differentiation is re- initiated when the heart tube starts looping, providing cells for the elongation of the heart
tube. These waves of differentiation are consistent with the two myocardial lineages ingressing sequentially through the primitive streak and thus reaching the heart region sequentially.
Myocardial sublineages. Further information about myocardial sublineages (Fig. 2) came from analysis of clonal contributions to the anterior skeletal muscles of the head and neck, which also express Actc1 during development. These anterior muscles, which derive from mesoderm present in the pharyngeal arches31 (Box 1), are clearly detectable by E14.5. Retrospective clonal analy-sis at E14.5 showed that masticatory muscles, derived from the mesoderm of the first pharyngeal arch, share
Nature reviews | Cardiology
Cardiogenic mesodermE7.25
Fetal heart ventricular wall
E13.5
Fetal heartE16.5
Cardiac crescentE7.5
Looped heart tubeE8.5
Chamber formationE10.5
Looped heart tube
Cardiac progenitors
Endoderm
LV
OFTRA
LA
RV
Primitivestreak
Pharyngealarches
Proepicardium
Cardiac crescent
PericardiumAp
Vp
Coronary artery
Coronary vein
Smooth muscle
ICV
Epicardium
PV
Coronary vessels
Pericardium
Endocardium
Myocardium
Epicardium
Endocardium
Myocardium
Valves
Trabeculations
Conduction system
Interstitial fibroblasts
Interstitial fibroblasts
IVS
LA
LVRV
Ao
SCV
RA PA
Fig. 1 | Cardiogenesis stages in the mouse. After gastrulation, cardiac progenitors migrate laterally towards the head folds at embryonic day (E) 7.25 (arrow), where the cardiac crescent forms (E7.5). At E8.5, the heart tube, which is composed of an inner layer of endocardium (yellow) and an outer layer of myocardium (orange), elongates from both the arterial (Ap) and venous (Vp) poles (arrows). At E10.5, cardiac chambers are visible as bulges from the heart tube. The proepicardium (E8.5) gives rise to the epicardium, which covers the myocardial surface and gives rise to interstitial fibroblasts (arrows), shown in the ventricular wall (E13.5). Coronary veins form superficially compared with coronary arteries. The different compartments and tissues of the heart are shown at E16.5. Ao, aorta; ICV, inferior caval vein; IVS, interventricular septum; L A , left atrium; LV, left ventricle; OFT, outflow tract; PA , pulmonary artery ; PV, pulmonary vein; RA , right atrium; RV, right ventricle; SCV, superior caval vein. Adapted with permission from reF.192, Cold Spring Harbor Laboratory Press.
www.nature.com/nrcardio
R e v i e w s
a common progenitor with the right ventricular myo-cardium, whereas facial expression muscles, derived from the second pharyngeal arch, are clonally related to the myocardium at the arterial pole, situated at the base of the aorta and pulmonary trunk (Fig. 3), with early left–right segregation of these clones32 (Fig. 2). Mesp1-based genetic tracing also showed clonality between
head muscles and the second lineage contribution to this arterial pole myocardium26. An extension of the retrospective clonal analysis to non- somite-derived skeletal muscles of the neck, which are formed from mesoderm of the posterior pharyngeal arches, showed a specific lineage relationship with the myocardium at the venous pole of the heart33. Here too, early left–right lineage segregation is observed, with the left or right tra-pezius muscles, for example, showing clonality with the myocardium of the left atrium, pulmonary vein and the left superior caval veins or with that of the right atrium and right superior caval veins, respectively (Figs 2,3). These analyses established an unexpected clonal rela-tionship between the left venous pole and the pulmonary trunk myocardium34, within the second lineage, as also indicated for inflow tract myocardium by clonal anal-ysis based on Mesp1 genetic tracing26. Fluorescent dye labelling of progenitor cells, followed by embryo culture, confirmed that some cells adjacent to the venous pole move anteriorly to contribute to the outflow tract myo-cardium35, with a left–right regionalization. This sub-lineage, which colonizes the dorsal regions of the atria, is distinct from the second pharyngeal arch sublineage that contributes to head muscles (Fig. 3). Outflow tract malformations are the most frequent form of congenital heart defects. More subtle anomalies at the venous pole, derived from the same lineage, can often go unnoticed, and checking for such defects at the time of corrective surgery can be life- saving36.
Lineage origins of later myocardial structures. Retrospective clonal analysis has provided additional insights into the clonal relationship between the venous pole and the atrioventricular canal22, also supported by dye- labelling experiments35. Analysis of clonal distribu-tions also indicates that the interventricular septum is composed of cells that originate from either the left or right ventricle30. Examination of labelled cells in com-pact versus trabeculated ventricular myocardium at E10.5 shows that these two types of myocardium are clonally related, with wedge- shaped clones, consistent with more cell proliferation in the compact layer than in the trabecular layer37. However, trabeculations in the fetal mouse heart have a polyclonal origin27. This find-ing is consistent with observations from precise recon-struction of the progression of trabeculation showing that trabeculation depends on the level of extracellular matrix and suggesting associated cell rearrangements38. Genetic tracing of the fetal myocardium (at E12.5), with the use of Hey2, Nppa and Sema3a markers, has shown that cardiomyocytes of the inner and outer domains of the ventricular wall maintain their respective positions in the postnatal heart, with an expansion of the compact myocardial layer during compaction of the ventricular wall39, consistent with the higher proliferative potential of the cardiomyocytes in the compact layer than those in the trabecular layer.
In addition to contractile cardiomyocytes, the heart contains cardiomyocytes with conductive properties that coordinate the contraction of the organ by prop-agating impulses from the pacemakers of the heart. Lineage studies have demonstrated that these two types
Box 1 | glossary of cardiogenesis terms
Coronary vasculatureThevascularsystemthatirrigatesthemyocardiumisbuiltupfromendothelialcellsthatmigrateintothedevelopinghearttoformanetworkofvessels,withrecruitmentofanouterlayerofsmoothmusclecells.Theimmaturevascularplexusdevelopswithoutbloodflow.Theimmaturevascularplexusiscomposedofsmall,similarlysizedvesselsorganizedinahighlybranchednetworkthatinitiatesonthesurfaceoftheventriclesandundergoesbranchingmorphogenesisandmassiveexpansionintotheentiremyocardium.Thevascularplexussubsequentlyextendsbranchestotheaortaandconnectswiththeaorticlumen,thusinitiatingperfusionofthecoronaryvessels.Bloodflowiscriticalinthesubsequentremodellingofthevesselsandthematurationintovesselswithsmoothmusclecoverage.
dorsal mesenchymal protrusionAtransitorystructureassociatedwiththedorsalmesocardiumoftheposteriorsecondheartfield,whichprogressivelyprotrudesventrallyintotheatrialcavity(fromembryonicday(E)10),whereitgivesrisetothemuscularbaseessentialforatrialandatrioventricularseptation.
Epithelial- to-mesenchymal transitionAprocessthatoccurswhenepithelialcellslosetheirpolarityandintercellularadhesiontodelaminate,undergoingmajormorphologicalandphenotypicchangestoacquirethemigratoryandinvasivepropertiesofmesenchymalcells.
gastrulationTheprocessbywhichthesingleepitheliallayeroftheearlyembryo,theepiblast,istransformedintoamultilayeredstructureascellsingressthroughtheprimitivestreak,withcharacteristicsofmesodermalprogenitorcellsaswellasendodermandectodermprogenitorcells.
Heart fieldTheembryologicalconceptofafieldreferstoapopulationofcellsataparticularlocationthatgiverisetoatissueororganduringdevelopment;thus,aheartfieldisasourceofcardiacprogenitorcells.
lateral plate mesodermSubdivisionofthemesodermaftergastrulationonthebasisofitspositionrelativetothemidlineoftheembryo.Theventralanddorsalaspectsofthelateralplatemesoderm,oneithersideofthecoelomiccavity,thusoverlyingtheendodermandtheectoderm,arenamedsplanchnicandsomaticmesoderm,respectively.Thelateralplate mesodermcontainsheartprogenitors,withpericardialcellsinthesomaticlayerandmyocardialandendocardialprogenitorcellsinthesplanchniclayer.Theseptumtransversumbisectsthecoelomiccavityintoacranialpericardialcavityandacaudalperitonealcavity.
Pharyngeal archesTransitorystructuresthatbulgeoutoneithersideoftheembryointhepharyngealregionandcontainamesodermalcorethatcontributestothepolesoftheheartandtocranialskeletalmuscles,surroundedbyendodermandectoderm.Thecardiacneuralcrestinvadestheposteriorarchesbeforemigratingintotheheart.
ProepicardiumAtransitorystructurethatconsistsofaheterogeneouspopulationofmesothelialcellsprotrudingfromtheseptumtransversumandlocatedadjacenttothevenouspoleoftheheartbetweenE8.5andE9.5.Cellsfromthisstructurespreadovertheoutersurfaceofthehearttoformtheepicardium.
Sinus venosusTermusedtodescribethevenousinflowtractoftheembryonichearttube,incontinuitywiththeatria,fromwhichvenouscomponentsdevelopandcardiacvascularcellsoriginate.Thesinusvenosusreceivesbloodfromthevitellineveinsconnectedtotheyolksac,theumbilicalveinsandthecommoncardinalveins.
R e v i e w s
Nature reviews | Cardiology
Box 2 | Cell lineage tracing methods
random retrospective clonal analysisThisapproachdependsonatemporallyandspatiallyspontaneouseventofcelllabelling(seefigurepanela).Amajoradvantageofthismethodisthatitisnon-invasiveandsystematic;thatis,theapproachdoesnotinvolveanypreconceptionaboutthepropertiesoftheprogenitorcell,suchasmarkergeneexpression.RetrospectiveclonalanalysishasbeenappliedinthemouseusingthenlaacZreportergenethatcarriesaduplicationthatintroducesastopcodonintothecodinglacZsequence,renderingitnon-functional.CloneswithafunctionallacZsequencearegeneratedrandomlyintimeandspaceasaresultofrare,spontaneousrecombinationthatremovestheduplication.Statisticalanalysisisrequiredtodemonstratethatthefrequencyofthelabellingpatternsisconsistentwithclonalanalysis.Thetimingoftherecombinationeventand,therefore,theoriginoftheclonecanbeapproximatelydeducedfromthenumberofcellsinthecloneandtheirfrequency.Forcelllineageanalysisofthemyocardium,nlaacZwastargetedtotheActc1locus,whichencodesactin,α-cardiacmuscle1(seefigure)22.lacZ+clones(blueoutlinedcells)expressβ-galactosidase(bluefilledcells)intheregionswheretheActc1promoterisactive(heartandheadmuscles).
genetic tracingConstitutive genetic tracing.Thisapproachtomonitorcelllineagesdependsontheidentificationofaspecificgeneticmarkeroftheprogenitorcellthatisnotexpressedatlaterstagesinitsprogenynorinothercelltypesorearlierprogenitors.Mostfrequently,constitutiveexpressionofCrerecombinaseisdrivenunderthecontrolofregulatorysequencesofthemarkergene,eitherasatransgeneorasatargetedallele,whichleadstotherecombinationofloxPsites,removalofaSTOPcassetteandexpressionofareporter(seefigurepanelb).Thisgenetictracingapproachpermanentlylabelscellsthathaveexpressedamarker.
Inducible genetic tracing or inducible clonal analysis.AninduciblesystemtoactivateCreinaprecisetimewindowreducestheriskofmisleadingresultsowingtoexpressionofthemarkergeneincellsatearlierorlaterstagesandallowstemporaltargetingofacellpopulation.Clonalanalysisispossiblebyusinglowdosesofinducer.FigurepanelcshowstheinducibleCre(CreER)–loxPsystem,inwhichexpressionoftheCrefusedtotheoestrogenreceptor(ER)isregulatedbyatissue-specificorcell-specificpromoter(greenoutlinedcells);hydroxy-tamoxifentreatmentinducesthetranslocationofCreERtothenucleus,leadingtoloxPrecombinationandexpressionofthereporter(greenfilledcells),drivenbyanubiquitouspromotersuchasRosa26.
Mosaic clonal analysisConfetti mice.TheconstructinConfettimiceincludesaseriesoffluorescentreportersofdifferentcolours,flankedbyloxPsites,whichcanberearrangedbytheactionofarecombinasetobepositionedundertheregulationoftheubiquitouslyexpressedRosa26locus.Forexample,expressionofatransgeneencodingreverse-tetracyclinetransactivator(rtTA)underthecontrolofMesp1canbeusedtolabelcardiacprogenitors(seefigurepaneld)26.Inthepresenceoftheinducerdoxycycline,rtTAbindstotetOsitesandactivatesCretranscription.Inductionoftherecombinasegeneratesstochasticmosaicsoflabelledcells,whichareclonallyrelatedifalowdoseofinducerisused.
Mosaic analysis with double markers.Inthemosaicanalysiswithdoublemarkers(MADM)ortwin-spotsystem,partialsequencesforfluorescentproteinsareseparatedbyloxPsites(seefigurepanele).Whenexposedtorecombinase,theexpressionofwhichiscontrolledbyregulatorysequencesofamarkergenesuchasMesp1(reF.29),thefluorescentreportersarereconstitutedandtheirexpressionisregulatedbyaubiquitouspromoter.Aftermitoticrecombination,greenandredfluorescentreportersdifferentiallylabelsistercellsasaresultofallelicsegregation.Afternon-mitoticrecombination,cellsaredouble-labelledandthusappearyellow.
C,C-terminaldomain;CFP,cyanfluorescentprotein;GFP,greenfluorescentprotein;N,N-terminaldomain;RFP,redfluorescentprotein;YFP,yellowfluorescentprotein.
Tissue-specificCre expression
activated by inducer(genetic tracing) or
low-dose inducer(clonal analysis)
Heart Head
nlacZ
β-Galactosidase
nlaacZ
Rare and spontaneousrecombination
a Retrospective clonal analysis d Mosaic clonal analyses with Confetti mice
b Constitutive genetic tracing
c Inducible genetic tracing or inducible clonal analysis
Ubiquitouspromoter
Inversion Inversion
Excision
Mitotic recombinationNon-mitotic recombination1
1 2
3
2
e Mosaic clonal analyses with MADM or twin-spot system
Actc1
Actc1
STOP
STOP Reporter
Cre
Tissue-specificpromoter
CreER
Tissue-specificpromoter
loxP N-GFP
STOP ReporterloxP
STOP nGFP
YFP
RFP
mCFP
loxP
Tissue-specificCre expression
1 2
1 2
3
Cre expression in cardiacprogenitors driven
by Mesp1–rtTA;tetO–Creand activated by low-dose
inducer
Ubiquitouspromoter
× ×
Mesp1-specificCre expression
nlaacZ
nlacZ
Rosa26
Rosa26
GFP-CN-RFP N-RFP
N-RFP
GFP-C
GFP-C
N-GFP
N-GFP
RFP-C
RFP-C
RFP-C
www.nature.com/nrcardio
R e v i e w s
of cardiomyocytes share a common origin. The central conduction system, including the atrioventricular node and the His bundle, segregates early during develop-ment from the cell precursors of the working myocar-dium40. By contrast, the peripheral conduction system is clonally related to contractile cardiomyocytes until E16.5 (reF.41). After segregation, cells of both lineages continue to proliferate, although to a lesser extent for the conductive cardiomyocytes. The peripheral conduction system has a dual origin that reflects the clonal bound-ary between the ventricles, such that the right ventricle and the right Purkinje fibre network are clonally distinct from the left bundle branch of the conduction system and the left ventricle40.
Therefore, the myocardium in the different cardiac regions has specific origins. The associated lineage deci-sions follow the main patterning events of the embryo, along the anterior–posterior axis and along the left–right axis, with an additional segregation into successive first and second lineages. In the next section, we discuss the molecular markers of these myocardial cell lineages, the concept of the first heart field (FHF) and second heart field (SHF) (Box 1), and the molecular regulation that governs the contribution of myocardial progenitors to different regions of the heart.
Genes that mark cardiac progenitorsA prerequisite to identify genes involved in lineage deci-sions is the characterization of genes that mark cardiac progenitor cells, which also provides tools for genetic tracing of cardiac progenitor derivatives in the heart. Gene expression can also pattern progenitor cell popu-lations and control their behaviour. Differences in gene expression point to heterogeneity within progenitor cell lineages.
Early cardiac progenitors within the primitive streak. As discussed previously, Mesp1-based genetic trac-ing labels most24, but not all (70%)23, cells in the heart. Furthermore, the Smarcd1 F1 enhancer is active in a subpopulation of Mesp1-expressing cells in addition to other cells in the primitive streak29. In both cases, clonal analysis within these expression domains identifies first and second lineages consistent with retrospective clonal analyses22,26,29. A clear distinction between the contributions of the two lineages on the basis of gene expression is challenged by a report published in 2017 that identified descendants of cells that have expressed Foxa2 in both the cardiac crescent, containing first lin-eage cells, and the medial domain of Isl1+ progenitors that contribute to second lineage derivatives of the heart42 (see Heart fields section). Foxa2 expression, which has been classically associated with endoderm, also occurs transiently in the anterior primitive streak and nascent mesoderm at the early and mid- streak stages43 as well as in the early gastrula organizer, which was shown in grafting experiments to provide cardiac cells44. Genetic tracing demonstrated that cells that have expressed Foxa2 at E5.75 (reF.45) or E6.5 (reF.42) contribute to a number of mesodermal tissues, including in the heart. Strikingly, cardiac labelling was reported to be restricted to the myocardium but exclusively the ventricles, both left and right, as well as to the endocardium and the epicardium42, whereas no such labelling is seen when induction takes place at E7.5. Analysis of chimeric embryos showed that Foxa2-mutant cells can no longer participate in cardiogenesis. These observations raise the question of the lineage identity of the Foxa2-expressing progenitors. Transcriptome analyses of sorted cells showed that Foxa2+ cells express lower levels of Mesp1 (reF.42). This finding is consistent with the possibility
First lineage Second lineage
FHF aSHF and PhA1 aSHF and PhA2 pSHF and PhA3–6
Right Left RightLeft
Masticatorymuscles
Right facialexpressionmuscles
Left facialexpressionmuscles
Left neckmuscles
Right neckmuscles
Skeletalmuscles
Chambermyocardium
RAiAVC
LAsAVC
RVRV LV LARAAVC
Aorta Pulmonarytrunk
Arterial polemyocardium
Venous polemyocardium
RSCVPVLSCV
Fig. 2 | Cell lineage tree of cardiac and associated skeletal muscles. Model of the lineage relationships between myocardial cells in the heart and skeletal muscles of the head and neck on the basis of data from retrospective clonal analyses in mouse models. Anterior–posterior patterning is highlighted in red, green and yellow. Left–right patterning is highlighted in dark and light blue. Specific contributions of first and second lineages are shown in bold. The dashed circle indicates the two sublineages that contribute to the pulmonary trunk myocardium. aSHF, anterior second heart field; AVC, atrioventricular canal; FHF, first heart field; iAVC, inferior AVC; L A , left atrium; LSCV, left superior caval vein; LV, left ventricle; PhA , pharyngeal arch; pSHF, posterior second heart field; PV, pulmonary vein; RA , right atrium; RSCV, right superior caval vein; RV, right ventricle; sAVC, superior AVC.
R e v i e w s
Nature reviews | Cardiology
that these cells were not detected in Mesp1-based clonal analyses (which did not identify interventricular clones), whereas retrospective clonal analysis had detected clones that span both ventricles at E8.5 (reF.22). By E12.5, cells
that had expressed Foxa2 represent ≥50% of ventricular cardiomyocytes, mainly concentrated in the ventricu-lar apex42. Whether these cells are distinct from those marked by Mesp1–Cre genetic tracing (which does not
A P
Fetal heart(E14.5)
LSCVPV
RSCV
LARA
LVRV
Ao
Looped heart tube(E8.5)Late headfold stage
(E7.5)
Heart tube
FirstPhA
SecondPhA
Cardiac crescent
First heartfield
Anterior andposteriorsecondheart field
Masticatoryhead muscles
Facialexpression
head muscles
Neural plate stage/early headfold stage
(E7.25–E7.5)
Gastrulation
Late streak stage(E7–7.25)
Mid-streak stage(E6.5)
R L
a
LV
RV
OFTPt
Sternocleidomastoid
Spino-trapeziusneckmuscles
Acromio-trapezius
Cardiogenic mesoderm
iAVC
sAVC
dLCARCA + vLCA
b
R L
LSCVPV
RSCV
dLAdRA
LVRV
AoPt
Fig. 3 | location of cardiac progenitors and their derivatives. a | First (red) and second (green) cell lineages and anterior (pale green and yellow) and posterior (dark green) subdomains of the second heart field are shown at different stages of heart and head development. Regions of the heart with a dual origin are shown with coloured dots. The progression of mesoderm formation during gastrulation is shown in orange. b | Representation of left (dark blue) and right (light blue) cell derivatives of the second heart field, also showing right and left skeletal muscles in the head and neck. Stages are as in panel a. Lateral views are shown at embryonic day (E) 6.5, E7, E7.25–7.5, E8.5 (in panel a) and E14.5, and frontal views are shown at E7.5 and E8.5 (in panel b). A , anterior ; Ao, aorta; dL A , dorsal L A ; dLCA , dorsal left common atrium; dRA , dorsal RA ; iAVC, inferior atrioventricular canal; L , left; L A , left atrium; LSCV, left superior caval vein; LV, left ventricle; OFT, outflow tract; P, posterior ; PhA , pharyngeal arch; Pt, pulmonary trunk; PV, pulmonary vein; R , right; RA , right atrium; RCA , right common atrium; RSCV, right superior caval vein; RV, right ventricle; sAVC, superior atrioventricular canal; vLCA , ventral left common atrium. Adapted with permission from reF.192, Cold Spring Harbor Laboratory Press.
www.nature.com/nrcardio
R e v i e w s
label all the ventricular myocardium23) remains to be shown. Foxa2 expression might mark another cardiac cell lineage or, alternatively, the Foxa2 gene might be transiently activated very early in subpopulations of the first and second lineage cells that subsequently locate anteriorly in both heart fields to contribute to ventricular myocardium. Clonal analyses of these cells should help to clarify their lineage.
Another example that points to an early difference during development in gene expression in cell subpop-ulations comes from analysis of Mef2c–AHF enhancer activity46. This enhancer marks second lineage progen-itors of the right ventricular and outflow tract myocar-dium. Activity of this enhancer driving the expression of a reporter transgene is already detectable at the late streak stage29, suggesting that these progenitor cells already differ at this stage from those of the second lin-eage that form the atrial and inflow tract myocardium. Therefore, distinct populations of mesodermal cells, marked by gene expression, seem to be specified at gastrulation, with regionalized contributions to the heart.
Heart fields. At the time when the cardiac crescent forms at about E7.5 (Fig. 1), two cell populations can be distinguished, those that are within the epithelial struc-ture of the crescent and those that lie medially to the crescent and also more posteriorly in splanchnic meso-derm (Box 1). These cells contribute to different regions of the developing heart, in accordance with the contri-butions of the first and second myocardial cell lineages. The sources of these cardiac cells are, therefore, referred to as the FHF and SHF47.
First heart field. Cells in the cardiac crescent constitute the FHF and mainly form the left ventricle3, as shown by dye labelling followed by embryo culture and by explant experiments in which cells that underwent myocardial differentiation expressed a left ventricular transgenic marker. The early heart tube, formed by fusion of the right and left halves of the cardiac crescent, has a mainly left ventricular identity. Gene markers of FHF progeni-tors have been hard to distinguish from genes expressed in cells in the cardiac crescent that are beginning to differentiate into cardiomyocytes. These genes include those encoding transcription factors, such as Gata4, Nkx2-5, Tbx5 or Hand1, many of which are implicated in myocardial cell differentiation but can also mark cardiac progenitors. Tbx5 expression marks the left ventricular myocardium, but Tbx5 is expressed dynamically in the early heart tube48 and marks a subpopulation of cells in the SHF49, therefore this gene cannot be considered to be lineage restricted.
Cardiac expression of Hand1 subsequently marks myocardial cells in the left ventricle, where Hand1− cells can compensate for the ablation of the Hand1+ cell population, illustrating cardiomyocyte cell hetero-geneity50. However, Hand1 cannot be regarded as a bona fide marker of a cell subpopulation of FHF derivatives because myocardial cells that have expressed Hand1 are also present in a domain of the outflow tract, which is entirely derived from the SHF, and Hand1 also marks the epicardium51.
Hcn4, which encodes a potassium- channel subunit, provides an interesting marker of the FHF52,53. Genetic tracing with a conditional Hcn4–CreERT2 cassette acti-vated at the early cardiac crescent stage showed classic first lineage labelling of the left ventricular myocardium and of some atrial myocardium. Endocardium was not labelled. Hcn4 transcripts are detectable at E6.75, before differentiating myocardial cells are present, and Cre induction at E6.0, which already shows some labelling in FHF derivatives in the heart, suggests that FHF myo-cardial progenitors are already marked by E6.5, when recombination and reporter activation should have occurred. Hcn4 continues to be expressed in differenti-ating myocardial cells of the cardiac crescent. Later, Hcn4 expression is re- activated in the conduction system of the maturing heart.
Another interesting marker is Sfrp5, which encodes a WNT decoy receptor54. This gene is expressed in the lateral caudal region of the FHF in undifferentiated cells, overlapping with Tbx5 transcripts. This pattern differs from Hcn4 transcripts, which seem to locate with differ-entiation markers at E7.5. Tamoxifen- inducible genetic tracing at E7.5 with an Sfrp5–CreERT2 shows that descendants of Sfrp5-expressing cells contribute to the left ventricular myocardium as well as to the atria. However, reporter induction at this and later stages also labelled myocardial cells in the outflow tract and venous pole, as well as the endocardium and epicardium, indicating that Sfrp5 is also activated in progenitors of the SHF.
Second heart field. The cardiac progenitor cells of the SHF express characteristic markers and are clearly distin-guishable by their location, medial to the cardiac crescent then dorsal to the heart tube, first in the dorsal mesocar-dium and then the pericardial wall, also extending anteri-orly and posteriorly in the pharyngeal mesoderm (Fig. 3). As the pharyngeal arches form, the SHF also contributes to their mesodermal core. The first identified marker of cardiac progenitors in pharyngeal mesoderm that con-tribute to the right ventricle and outflow tract of the heart was a transgene that had inserted into the Fgf10 locus2. In this early study, dye labelling followed by embryo cul-ture confirmed the contribution of mesoderm present in the second pharyngeal arch to the development of the outflow tract myocardium. Fgf10, together with Fgf8, are expressed by cells in the anterior SHF from the cardiac crescent stage. These genes encode fibroblast growth fac-tors (FGFs) that are implicated in progenitor cell prolif-eration and are required for the correct formation of the right ventricle and outflow region of the heart55.
Isl1 expression first gave an indication of the full extent of the SHF, extending more posteriorly than that of Fgf10 and Fgf8, with Isl1-mutant mice showing defects at the venous pole in the atria as well as at the arterial pole in right ventricular and outflow tract devel-opment56. Genetic- tracing experiments show that cells that had expressed Isl1 contribute to the endocardium as well as the myocardium. Similar to other ‘cardiac markers’, Isl1 expression is present at a number of sites in the embryo, including the endoderm and the cardiac neural crest, which can complicate the interpretation of genetic- tracing experiments57. More sensitive genetic- tracing
R e v i e w s
Nature reviews | Cardiology
experiments have detected Isl1 expression in the FHF58, as well as the SHF, although this low- level expression does not seem to be functionally relevant for FHF deriv-atives. A few left ventricular cells are also labelled with Isl1-based genetic tracing59.
Other genes encoding transcription factors that are expressed widely in the SHF include Nkx2-5, which encodes the homeobox protein NKX2-5 that has a major role in cardiomyocyte differentiation and is also expressed in the cardiac crescent. In the cardiac progen-itor cells of the SHF, NKX2-5 has a distinct function in regulating the maintenance of the progenitor cell pool60. Nkx2-5-mutant mice lack an outflow tract and have other embryonic heart defects, with lack of ventricular compartmentation61.
A classic marker of the anterior SHF is provided by the activity of the Mef2c–AHF enhancer46. Transgenes under the control of this enhancer are expressed in SHF progenitors, which genetic- tracing experiments show contribute to the right ventricular and outflow tract myocardium46. Expression of other transcription fac-tor genes, such as Foxc1 and Foxc2, is also present in anterior SHF progenitors, and in the absence of these transcription factors, the outflow tract myocardium does not form correctly62. A conceptually interesting aspect of marker gene labelling of anterior SHF derivatives comes from the analysis of Six2-expressing progenitor cells63. Controlled, temporal genetic tracing (Box 2) with an inducible Six2–EGFP–CreERT2 line showed that Six2 activation dynamically marks successive subpopulations of progenitors that contribute to the developing right ventricle and then to the outflow tract, in contrast to genes that are thought to be continuously expressed in the SHF progenitor cell pool.
Some genes expressed in the anterior SHF, such as Tbx1 (reviewed previously64), mark only a subset of pro-genitors. Tbx1 marks cells that contribute to the inferior part of the outflow tract that subsequently forms the myocardium at the base of the pulmonary trunk65, which is particularly affected in Tbx1-mutant hearts66. Expression of Sfrp5 in outflow tract progenitor cells54 does not overlap with that of Tbx1, suggesting that Sfrp5 marks a complementary population of myocardial cells, such as at the base of the aorta, as seen with marker transgenes67. In Tbx1-mutant mice, formation of skeletal muscles in the head is also affected (a detailed discussion of Tbx1 in cardiac development has been published pre-viously68). These findings suggest that Tbx1-expressing cells belong to the myocardial sublineage that gives rise to head and heart striated muscles. Unexpectedly, how-ever, Tbx1-mutant mice also have defects at the venous pole of the heart49. This finding is reminiscent of the cell sublineage that contributes to the pulmonary trunk and the left venous pole, and indeed the transcription fac-tor encoded by Tbx1 also has a functional role in non- somitic neck muscles derived from this sublineage33. This cell sublineage is distinct from the progenitors that give rise to head muscles and myocardium at the arterial pole (Fig. 2), indicating that Tbx1 expression is probably independently activated in cells derived from both sub-lineages. In the absence of Tbx1, reduced proliferation and premature differentiation are observed in cells in the
anterior SHF, as well as modifications to the morphology and epithelial properties of these cells69. In Tbx1-mutant mice, a striking change also occurs in the expression domains of anterior SHF markers such as Fgf10, which is shifted more posteriorly. Posterior SHF markers are also perturbed.
Tbx5 is a marker of the posterior SHF, and genetic- tracing experiments show that SHF cells expressing Tbx5 contribute to atrial myocardium and to the myocardium of the venous pole of the heart but not to the arterial pole29. As discussed above, Tbx5 is also expressed in the FHF and its derivatives, in the cardiac crescent and later in the left ventricle. Other marker genes expressed in subdomains of the posterior SHF include Osr1 (reF.70) and Wnt2 (reF.71), which are required for aspects of venous pole development. Tbx18 expression marks a posterior population of cells situated laterally, out-side the Isl1 or Nkx2-5 expression domain of the SHF. These cells contribute to the caval vein myocardium, which continues to express Tbx18, which is shown to be required for the formation of this myocardium. This pattern of expression is also seen for Sfrp5 (reF.54). This Tbx18+ progenitor population was proposed to consti-tute a third heart field72,73. However, this distinction on the basis of gene expression does not equate to findings from cell lineage analysis, which showed that caval vein myocardium is not derived from a specific cell sublin-eage but shares common cell progenitors with the atrial myocardium34 (Fig. 2).
In addition to the striking anterior–posterior region-alization of marker gene expression in the SHF, which depends on an underlying gradient of Hox gene expres-sion, regulated by retinoic acid74, left–right regional-ization, under the influence of Nodal signalling, also occurs. Nodal signalling leads to Pitx2 expression in cells of the left side of the SHF75. Dye labelling followed by embryo culture showed the contribution of these left posterior SHF cells to the dorsal left atrium and superior atrioventricular canal35. Explant experiments, with the use of transgenic left–right markers of atrial myocardium, showed that posterior SHF cells already have left–right atrial identity1. Left–right signalling and Pitx2c expression also pattern the outflow tract myocar-dium and the subsequent morphogenesis of the base of the great arteries76,77.
Molecular regulation of cardiogenesisMolecular regulation of progenitor cell behaviour operates at the level of gene regulatory networks. This molecular regulation of cardiogenesis also involves protein interactions between transcription factors and cofactors that modulate their specificity. Interactions with components of chromatin and its 3D organiza-tion in the nucleus determine the accessibility of genes for transcriptional regulation. Long non- coding RNAs (lncRNAs) add another level of regulation.
Regulatory networks in the SHF. The SHF receives signals from surrounding tissues, such as sonic hedge-hog (SHH) from the endoderm78, and the SHF is also a source of cell autonomous signalling, as seen for FGF55. In general, canonical WNT, FGF, Hedgehog and Notch
www.nature.com/nrcardio
R e v i e w s
signalling pathways promote proliferation in the SHF and thus ensure maintenance of the cardiac progenitor cell pool, whereas bone morphogenetic protein (BMP) and non- canonical WNT signalling are implicated in entry into the cardiomyocyte differentiation pro-gramme79,80. The antagonism between pro- proliferation and pro- differentiation signalling is illustrated by the role of HOPX81, a transcriptional cofactor expressed in cardiomyocyte progenitor cells just before their dif-ferentiation. HOPX forms a repressive complex with SMAD4 — an interaction that is activated by BMP signalling — and HDACs that binds to and downregu-lates the transcription of genes in the canonical WNT pathway81. In the absence of HOPX, cardiomyocyte dif-ferentiation is delayed. Thus, at the distal border of the outflow tract there is a transition zone where BMP sig-nalling is high, cell proliferation is reduced and cardiac muscle begins to form.
In the SHF, the interplay between signalling path-ways and regulatory genes has an important role in determining progenitor cell behaviour and contribu-tion to the heart. An extensive literature has addressed this aspect of cardiogenesis. Here, we present two examples that illustrate these gene regulatory networks (Fig. 4). In the anterior SHF, GATA4 and FOXC2 reg-ulate Isl1 expression82,83 (Fig. 4a). ISL1 directly activates the enhancers of Mef2c84 and Fgf10 (reF.85) that drive expression in SHF progenitor cells, and Isl1 lies upstream of Fgf8 in the gene network56. TBX1 also directly acti-vates the Fgf10–SHF enhancer85 and lies upstream of Six1 and Eya1 in the gene network, which directly regulate Fgf8 expression86. NKX2-5, which is present at lower levels in the SHF than in cardiomyocytes owing to partial repression by an ISL1–HDAC complex87, also participates in this network as a direct activator of the Fgf10–SHF enhancer85. NKX2-5 also dampens
BMP2 activity, which feeds back positively on Nkx2-5 expression60,88,89. NKX2-5 together with FOXH1 also directly activate a Mef2c–SHF enhancer90. By contrast, in differentiating cardiomyocytes in the heart, NKX2-5 directly represses Fgf10 and Isl1 expression60,85,91. Isl1 downregulation in the heart also reflects the hypoxic environment of the heart tube compared with the nor-moxic conditions of the SHF, leading to high levels of HIF1α, which forms a complex with HES1, SIRT1 and TBX5 on a regulatory region of Isl1 to directly repress Isl1 transcription87 (Fig. 4a).
In the posterior SHF, TBX5 and Hedgehog sig-nalling92 are required for the formation of the dorsal mesenchymal protrusion (DMP) (Box 1), which has an important role in atrial septation93 (Fig. 4b). Genetic- tracing experiments show that the DMP is derived from SHF cells that had expressed Isl1 and Tbx5 (reFs92,94). The DMP is also labelled with the use of a Cre transgenic line driven by the Mef2c–AHF enhancer95, suggesting an overlapping contribution from anterior as well as pos-terior SHF cells. Temporally controlled genetic- tracing experiments show that SHF progenitor cells that have expressed the effector of the Hedgehog pathway, GLI1, contribute to the atrial septum, including that formed from the DMP96. TBX5 directly activates the expression of Cdk6, which encodes a cyclin- dependent kinase, thus promoting cardiac progenitor cell proliferation92. TBX5 also activates expression of Osr1 (reF.92), which encodes a transcription factor required for the formation of the atrial septum. Activated Hedgehog signalling res-cues atrial septal defects in Tbx5-mutant embryos, and this effect is at least partly due to direct activation by TBX5 of Gas1 expression92, which encodes a Hedgehog signalling component. Downstream target genes in the Hedgehog pathway in the posterior SHF include Foxf1a (also known as Foxf1) and Foxf2, which encode forkhead
Mesp1 Directed migration(Rasgrp3)
EMT from the primitivestreak (Snai1)
Cardiovascular specification(Hand1 and Gata4)(Etv2 and Pdgfra)
Epiblast(Cdh1)
Pluripotency(Nanog and Pou5f1)
Foxc2, Gata4 Isl1 Fgf10 Tbx1
Nkx2-5 Bmp2, Smad1
Hif1a, Hes1, Sirt1
Mef2c–SHFenhancer
Fgf8
Six1, Eya1
Osr1 Cdk6
Tbx5 Gas1 Hedgehog pathway
Gli3, Gli1
Ccnd2, Cdk4
Foxf1a Foxf2
a Anterior SHF
c
b Posterior SHF
Gata4
Fig. 4 | gene regulatory networks in cardiac development. a | A regulatory network in the anterior second heart field (SHF), centred on Isl1, Tbx1 and Nkx2-5, that affects the development of the arterial pole in mice. Red lines indicate inhibition in cardiomyocytes. b | A regulatory network in the posterior SHF, centred on Tbx5, Gata4 and the Hedgehog signalling pathway , that affects the development of the dorsal mesenchymal protrusion and subsequent atrial septation in mice. c | Mesp1-dependent genes and developmental processes implicated in cardiogenesis. Dashed lines indicate that activation might be indirect. EMT, epithelial- to-mesenchymal transition.
R e v i e w s
Nature reviews | Cardiology
transcription factors functionally important for atrioven-tricular septation97. Foxf1a expression not only is directly regulated by the Hedgehog pathway factors GLI1 and GLI3 but also is a direct target of TBX5. The transcrip-tion factor GATA4, which is present in the posterior SHF, also intervenes in this regulatory network, directly activating Gli1 expression and promoting cell cycle pro-gression in SHF progenitors98. Atrial and atrioventricu-lar septal defects in Gata4-mutant mice are rescued by constitutive Hedgehog signalling in the SHF (Fig. 4b).
Interactions between transcription factors. Essential cardiac transcription factors such as GATA4, NKX2-5 and TBX5 can interact with each other99,100, providing a further degree of network complexity, with impor-tant implications also for the severity of combinations of mutant phenotypes of relevance to human con-genital heart disease. Gata4 and Tbx5 interact genet-ically101, and modelling in induced pluripotent stem cells of human GATA4 mutations shows how cardiac identity is conferred by binding of GATA4 and TBX5 and the mediator complex to superenhancers; mutant GATA4 proteins do not bind to the superenhancers and thus TBX5 occupancy at the superenhancer is lost102. Detailed functional analysis of GATA4, NKX2-5 and TBX5 shows the extent of these cooperative interac-tions across the genome, with the crystal structure for TBX5–NKX2-5 binding to a DNA target demonstrating the atomic basis for this interaction103. Interdependent genome occupancy of cardiac regulatory sequences reduces the probability of inappropriate gene activa-tion owing to off- target effects, seen for example with NKX2-5 mutants that interact excessively with ETS transcription factors, leading to off- target binding and aberrant gene activation104.
Transcriptional cofactors. Cofactors that form a com-plex with cardiac transcription factors and modulate their activity are also an important facet of network regulation. This concept is illustrated by proteins that bind to ISL1. Ajuba represses ISL1 transcriptional activity in a retinoic- acid-dependent manner such that in the absence of ajuba, the ISL1+ progenitor cell pool expands, leading to excessive numbers of cardio-myocytes at the venous and arterial poles105. LDB1 also forms a complex with ISL1, protecting ISL1 from degradation and promoting 3D chromatin organization around ISL1 target genes106.
Regulation of chromatin accessibility. Interaction between cardiac transcription factors and chromatin components is an indication of their ‘pioneer’ function, as in the case of TBX1, which binds to ASH2L107 (a core component of the histone methyl transferase complex) and interacts at the chromatin level with mixed- lineage leukaemia (MLL; also known as KMT2A) family members to regulate monomethylation of histone 3 lysine 4 of target genes108. In addition, TBX1 binds to BAF60A (also known as SMARCD1), a component of the SWI–SNF complex, recruiting it to chromatin109. In this way, regulatory sequences are primed for access to other cardiac transcription factors. Another subunit
of the SWI–SNF complex, BAF60C (also known as SMARCD3), marks early cardiac progenitors in the primitive streak and is present later in the SHF29. A combination of TBX5 with BAF60C and GATA4, which activates Nkx2-5 expression, can induce beat-ing myocardial tissue when ectopically expressed in embryonic mesoderm110.
In addition to chromatin per se, the 3D regulation of genome organization within the nucleus is another aspect of cell fate control. In this context, in cardiac pro-genitor cells, cardiomyocyte- specific genes are retained in a repressive state at the nuclear lamina, probably anchored by HDAC3, and these regions translocate from the periphery to the centre of the nucleus at the onset of cell differentiation111.
Non- coding RNAs also participate in the regulation of cardiac progenitor cells. The lncRNA Braveheart is present early during cardiac development, potentially acting upstream of MESP1 at gastrulation. Braveheart interacts with SUZ12, a component of the Polycomb complex, to relieve epigenetic repression on cardiac regulatory genes112. At the structural level, the func-tion of this lncRNA depends on a G- rich internal loop motif113. Fendrr is another lncRNA that is encoded in the Foxf1 locus. Fendrr interacts with members of the Polycomb complex and with the Trithorax–MLL complex to modulate repression, as well as promoting activation, of regulatory genes in cardiac progenitor cells114.
The molecular circuitry discussed here provides a basis for the patterning of cardiomyocyte progenitors and potentially underlies the segregation of myocardial cell sublineages. In the next section, we discuss non- myocardial cell lineages and their derivatives in the heart, with a focus on the mesodermal cell lineage of the endocardium and epicardium.
Non- myocardial cell lineagesAll the components of the heart are of mesodermal ori-gin with the exception of neural crest cells14,115,116, which derive from neurectoderm and migrate into the arterial pole of the heart where they have an important role in septation and outflow tract morphogenesis. Neural crest cells contribute to smooth muscle derivatives and mesen-chymal cells that will form the heart valves. In addition, the parasympathetic nervous system of the heart derives from the cardiac neural crest, whereas the sympathetic innervation derives from trunk neural crest cells117.
Grafting experiments have shown that progenitor cells for the myocardium, endocardium and pericardium are present in the same region of the epiblast21. When these progenitor cells ingress through the primitive streak, they express Mesp1 (reF.118) and later share mark-ers such as expression of Isl1 or Nkx2-5 (reFs56,59,61). In the cardiac crescent, the different cardiac cell types can be distinguished histologically: squamous pericardial cells are part of the somatic lateral plate mesoderm, whereas myocardial cells are found in the epithelial splanchnic lateral plate mesoderm and endocardial cells appear as small clusters ventrally, which delaminate from the splanchnic mesoderm20.
The contribution of Mesp1-expressing progen-itors not only to the myocardium but also to other
www.nature.com/nrcardio
R e v i e w s
cardiovascular cell types makes it possible to extend the cell lineage analysis and to look for the presence of multipotent progenitor cells, as suggested by clonal analysis of cardiac progenitors derived from pluripotent stem cells in vitro119–121. By clonal analysis in vivo26, first lineage progenitors were found to be unipotent, contributing to either myocardial or endocardial cells. Second lineage myocardial progenitors were unipotent or bipotent for cardiomyocytes and endocardial cells or for cardiomyocytes and smooth muscle cells, notably at the poles of the heart. Labelled cells were also detected in the epicardium, mainly after clonal induction at an ear-lier time point (E6.25). Most clones in the epicardium were unipotent, with rare, bipotent clones also showing myocardial labelling. This finding suggests that most epicardial progenitors arise from an independent pop-ulation of Mesp1-expressing progenitors. Analysis of twin- spot derivatives in mosaic mice showed the pres-ence of bipotent clones contributing to the myocardium and endocardium or to endothelial cells and smooth muscle cells, in addition to the majority of clones that were unipotent, either myocardial or endothelial29.
Endocardium. The endocardium, which constitutes the inner endothelial layer of the heart, develops in close association with the myocardium, providing a scaffold around which differentiating myocardial cells assem-ble (reviewed previously122). Signalling between these two cell types is an important aspect of cardiogenesis, with the endocardium regulating critical morpholog-ical events such as trabeculation38. Early mesodermal markers such as Kdr are expressed by progenitor cells of both the endocardium and myocardium in the prim-itive streak123,124 as well as by other cell lineages, nota-bly haematopoietic progenitors. The common origin of endocardial and myocardial cells is also illustrated by genetic- tracing experiments with heart field markers such as Isl1 (reF.56) or Nkx2-5 (reF.61). Tie2, which is expressed in the embryo later than Kdr, is specific to the endothelium, including the vasculature throughout the embryo and the endocardium124, whereas Nfatc1 expres-sion specifically marks cells of the endocardium125–127. Cell lineage studies, mainly based on genetic tracing, have shown that the endocardium is an important source of endothelial cells of the coronary vasculature (Box 1). The endocardium also gives rise to most mes-enchymal cells that constitute the cardiac cushions and subsequently contribute to the heart valves. A subpopu-lation of Nkx2-5+ endocardial cells, present mainly in the outflow tract and dorsal atria from E8, shows transient haemogenic activity (that is, the capacity to differentiate into haematopoietic progenitors)128. Thus, the mouse embryo has traces of the cardio–vasculo haemato poietic cell lineage, which has been extensively documented for the zebrafish129.
Epicardium. The epicardium is a major source of non- myocardial cell types in the heart (Fig. 5). At E8.5, epi-cardial progenitors are present in the proepicardium130, a transitory cluster of mesothelial cells protruding from the septum transversum, in proximity to the sinus veno-sus of the early heart tube (Box 1). These cells attach and
grow over the outer surface of the heart from about E9.5 to form an outer epithelial layer, the epicardium. From E12.5, cells delaminate by a process of epithelial- to-mesenchymal transition (EMT) (Box 1) and enter the underlying myocardium131. Both the proepicardium and the subsequent epicardium are marked by Isl1-based and Nkx2-5-based genetic tracing59,132, but these genes also mark other cardiac cells. Tbx18 and Wt1 are more specifically expressed by epicardial cells. Genetic trac-ing with Cre expression driven by regulatory sequences of these genes suggested that the epicardium gives rise to some myocardial cells133,134. However, because Tbx18 and Wt1 expression can be detected in some cardio-myocytes30,135,136, or in potential cardiomyocyte proge-nitors137,138, this interpretation has been controversial. The epicardium is the major source of smooth muscle cells of the coronary vasculature and contributes to most of the fibroblast population of the heart.
Interstitial fibroblasts of the heart. Classic embryolog-ical experiments with avian embryos and subsequent genetic- tracing experiments in the mouse embryo have shown that interstitial cells of the heart derive from the epicardium (Fig. 5). The epicardium is the main source of cardiac fibroblasts (reviewed previously139). Fibroblasts that form the annulus fibrosus140,141 — the fibrous tissue that separates the atrial and ventricular chambers — are also of epicardial origin, which is also the case for the fibroblast contribution to the leaflets of some of the heart valves131. Cardiac fibroblasts close to the atrioventricular junction in the interventricular septum are of endocar-dial origin. The neural crest contributes to a small subset of fibroblasts in the right atrium142,143 (Fig. 5e).
One of the problems with tracing cardiac fibroblasts has been the lack of specific markers for this interstitial cell population. However, the definition of fibroblast markers and combinations of markers now makes their identification more reliable139. Heterogeneity in the car-diac fibroblast population has also been regarded as a complication in defining this cell population. Mouse lines such as Pdgfra–GFP and Col1a1–GFP are reliable markers of cardiac fibroblasts, derived either from the endocardium or epicardium, and of cardiac fibroblasts after activation143. Tcf21, expressed in the epicardium where it is required for lineage- specific EMT of cardiac fibroblasts, also remains a marker of these cells in the heart144. Another marker, the extracellular matrix pro-tein periostin, is produced by fibroblasts in the devel-oping heart but not in quiescent fibroblasts in the adult heart. After injury, myofibroblasts that express Acta2 (encoding α- smooth muscle actin) and Postn (encod-ing periostin) proliferate and have an important role in scar formation and adaptive remodelling of the heart139. Genetic tracing with an inducible Tcf21–MerCreMer shows that Postn- expressing cells are derived from the Tcf21-expressing fibroblast population145.
Vascular smooth muscle cells. Smooth muscle cells of the coronary vasculature (Fig. 5), marked by Pdgfrb expression, are also predominantly derived from the epicardium (reviewed previously146), as indicated by genetic tracing with markers such as Tbx18 (reF.133).
R e v i e w s
Nature reviews | Cardiology
Mesenchymal progenitor cells, initially present in the cushions of the atrioventricular canal and the out-flow tract, subsequently give rise, after short- range migration, to vascular pericytes and smooth muscle cells, mainly in the ventricular septum and to a lim-ited extent in the ventricles147. These mesenchymal progenitor cells derive from the endocardium, indicat-ing an unexpected endothelial origin for this vascular smooth muscle. In addition, neural crest cells, which migrate into the arterial pole, contribute to coronary artery smooth muscle, notably within the ventricular septum116. Another smooth muscle layer surrounds the arterial pole at the base of the great vessels. This cell lineage is associated with the neural crest, as well as the myocardium, forming from mesodermal cells in the same location as the myocardial progenitors of the second lineage. Genetic tracing with the neural crest
marker Wnt1, compared with the use of Six2 or the Mef2c–AHF enhancer that mark myocardial progeni-tors (Fig. 5c), has shown little overlap between the two cell sources46,63. Smooth muscle not only at the arterial pole but also at the venous pole of the heart corre-sponds to labelling seen from the myocardial–smooth muscle bipotent progenitor cell population reported in clonal analyses26.
Vascular endothelial cells. The epicardium was ini-tially also thought to be the source of endothelial cells of the coronary vasculature, on the basis of retro viral labelling in the chick148. However, genetic- tracing experiments in the mouse embryo have challenged this view, and other cell sources have been identified149,150 (Fig. 5). Genetic tracing with epicardial markers, such as Tbx18, did not label coronary endothelium151. Following
Neural tube
Pericardial cells
Endocardial cells
Myocardial cells
A P
Proepicardium
Epicardialcells
ContractilecardiomyocytesConductivecardiomyocytes
Cardiac neuralcrest cells
OFT and coronarysmooth muscle cells*
Cardiogenicmesoderm
Endocardialcushions
Endocardialcells
Coronaryendothelial cells*
Parasympatheticneurons
a E7.5
b E9.5
Cardiacfibroblasts*
Cardiacfibroblasts*
Cardiacfibroblasts*
Coronary smoothmuscle cells*
Coronary smoothmuscle cells*
Lymphatic vessels
Mesenchymalcells of valves*
Mesenchymalcells of valves*
Mesenchymalcells of valves*
OFT smoothmuscle cells*
YS
SV
Haematopoieticcells
SHF-derivedNeural-crest- derived
c Distal outflow tract d Coronary endothelium e Cardiac fibroblasts
Overlapping contribution
Aorta Dorsal side of the heart Ventral side of the heartPulmonary trunk
Endothelium
Adventitia
Smoothmuscle cells
Epicardium-derived
Neural-crest-derived
SV-derived
Endocardium-derived
Endocardium-derived
SV
Fig. 5 | Sources of the different cardiac cell types. a | Most cardiac cells originate from the cardiogenic mesoderm. b | Later contributions from the sinus venosus (SV) and the proepicardium (in blue) are shown, as well as from cardiac neural crest cells, which originate in the neural tube and migrate through pharyngeal arches 3–6 (in purple). Overlapping lineages are marked with an asterisk. Minor contributions are indicated with a dotted line. c–e | Regionalization of the distinct cell sources colonizing the distal outflow tract (OFT) or coronary endothelium, or giving rise to the interstitial cardiac fibroblasts. A , anterior ; E, embryonic day ; P, posterior ; SHF, second heart field; YS, yolk sac. Part d adapted with permission from reF.156, Elsevier. Part e adapted with permission from reF.193, Elsevier.
www.nature.com/nrcardio
R e v i e w s
observations of heterogeneity in the proepicardial cell population, genetic tracing with the Scx and Sema3d markers labelled a few (5–10%) endothelial cells of the coronary vasculature152,153. By contrast, studies of angio-genic sprouting from the sinus venosus151,154, coupled with genetic tracing using the Aplnr–Cre mouse line152 and multicolour clonal analysis151, identified another major cell source of coronary endothelium. Venous endothelial cells derived from the endocardium at the sinus venosus were shown at E11.5 to dedifferentiate as they migrate and invade the embryonic myocardium, where these cells then redifferentiate into endothelial cells of the arteries and capillaries, whereas cells on the surface of the heart contribute to the endothelial cells of coronary veins151,155. This process depends on VEGFC and the peptide hormone apela (also known as elabela), both produced by the epicardium152,156. The sinus venosus cell lineage, which emerges dorsally and basally, spreads towards the apical and lateral sides of the ventricles (Fig. 5d).
In a complementary pattern, the endocardium was shown to contribute coronary endothelial cells mainly in the interventricular region and ventral wall of the ventricles. Extensive labelling of endothelial cells of arteries and capillaries, but not veins, was shown by genetic tracing with Cre expression driven by regula-tory sequences of Nfatc1, which is expressed mainly in endocardial cells152,157. This endocardial- to-vascular endothelial lineage deployment depends on VEGFA signalling from the myocardium, via the VEGFR2 receptor157,158, and not on apela–apelin receptor sig-nalling156. Subsequent genetic tracing with Npr3, which is more specifically expressed in the endocar-dium (but not in the sinus venosus endocardium) than Nfatc1, indicates a substantial but smaller contribu-tion of endocardium- derived vascular endothelial cells to the interventricular septum and the adjacent ventricular wall159.
The possible functional relevance of spatially spe-cific locations of vascular endothelial cells of different origins remains to be investigated. One hypothesis is that the sinus venosus sprouts in response to devel-opmentally timed signals, whereas the endocardium is stimulated by the physiological cue of hypoxia, via the hypoxia- inducible gene Vegfa. Intriguingly, genetic tracing with Cre expression under the control of reg-ulatory sequences of the proepicardial marker genes Sema3d and Scx also labelled endothelial cells of the sinus venosus or traced to cells in the endocardium, respectively, suggesting that some proepicardial- derived endothelial cells transit through these major sources of coronary endothelium. The interpretation of this result requires further investigation given the more distant clonal relationship between epicardial and endocardial progenitors26. The contribution of the proepicardium to ventricular endothelial cells, notably of the arteries, and the morphogenesis of this coro-nary vasculature continues to be an issue, as shown by genetic- tracing and gene- deletion experiments with Cre expression driven by the Gata4–G2 enhancer, which marks an epicardial progenitor cell population distinct from that expressing Wt1 (reF.160).
Lymphatic vessels of the heart. The lymphatic vascula-ture of the heart forms along the coronary vasculature, thus entering the heart slightly later than the coronary vasculature, at E12.5, from both the sinus venosus and the outflow tract. Genetic tracing with the Tie2 or Pdgfb markers, together with the observation of sprouts from the cardinal vein, adjacent to the sinus venosus, indi-cated an endothelial origin of the lymphatic vessels161. However, only 80% of lymphatic vessels were labelled. Genetic tracing experiments with markers of the haemo-genic endothelium have suggested another origin in the yolk sac161, although the specificity of the markers is debated162.
Cardiac cushions and valve formation. Heart valves have a critical role in controlling the direction of the blood flow in the heart. Valve formation is initiated in the embryonic heart by delamination of endocardial cells by EMT — as shown in pioneering experiments in the chick163,164 — specifically in the outflow tract and atrioventricular canal, in response to signals from the underlying myocardium165, such as BMP2 as shown for the mouse embryo166. Chamber endocardium does not have this potential to undergo EMT. These endocardial- derived cells form the cardiac cushions from E9.5 in the mouse. Genetic- tracing experiments with Tie2–Cre showed that valves are predominantly derived from the endocardium167. Later during cushion development, epicardial- derived fibroblasts, referred to as valvular interstitial cells in this context, also contribute to valve mesenchyme, as indicated by genetic- tracing experi-ments with Wt1–Cre131, specifically in the mural leaf-lets of the atrioventricular (mitral and tricuspid) valves, whereas the septal leaflets remain predominantly endocardium- derived in these valves.
In contrast to atrioventricular valves, the aortic and pulmonary semilunar valves receive a contribution from cardiac neural crest cells. Genetic tracing based on markers of these cells (Cx43–lacZ168, Wnt1–Cre116 and Krox20–Cre169) showed colonization of the left and right leaflets of semilunar valves. The semilunar valves also have a contribution from the endocardium170. However, the posterior, non-coronary leaflet of the aortic valve and the anterior leaflet of the pulmonary valve are poorly col-onized by neural-crest-derived and endocardium-derived cells, sug gesting a distinct origin. This origin has now been traced with Tnnt2–Cre, which results in labelling of intercalated cushion cells, which give rise to valve leaf-lets, in the wall of the cardiac outflow tract171,172. These cells retain signatures of previous activation of the cardiac troponin promoter, although they do not have a myo-cardial phenotype. These cells are not derived from diff-erentiated myocardial cells but arise instead from direct differentiation of SHF progenitors into valve cells.
The origins of the non- myocardial cells of the heart and the relative contributions of the epicardium and the endocardium are now better understood. Genetic trac-ing has mainly provided the basis for these conclusions, some of which, such as a possible myocardial contri-bution of the epicardium, have been controversial. The specificity of Cre drivers, as well as their availability, is a limiting factor. As more precise information on gene
R e v i e w s
Nature reviews | Cardiology
expression becomes available, novel tools for genetic tracing and adaptations for clonal analysis (Box 2) should clarify some of the remaining issues.
Convergences in the cell lineage treeStrikingly, several cardiac cell types, including fibro-blasts, coronary endothelial cells, and coronary smooth muscle cells, have multiple sources (Fig. 5a,b), indicating convergences in the lineage tree. When specific mark-ers have been identified, genetic- tracing experiments have shown that the different cell sources tend to colo-nize specific regions of the heart (Fig. 5c–e). This obser-vation raises the question of whether cell lineages can compensate for each other when one is lost or reduced. This hypothesis has been elegantly demonstrated in the case of the coronary endothelium, where the main cell sources can be traced with either Nfatc1 or Aplnr as genetic markers. When sinus venosus angiogenesis is deficient, such as in mice with mutations in Aplnr or Ccbe1 (which encodes a component of VEGFC sig-nalling), endocardial- derived vessels expand, leading to normal adult cardiac function156. Activation of the endocardium was proposed to be induced by hypoxia and is marked by the upregulation of the transcrip-tion factor SOX17. Therefore, overlapping cell lineages might have specific signalling driving their deployment. Compensation of the absence of one lineage by an alternative cell progenitor source confers robustness on heart development.
Divergences in the cell lineage treeThe reconstruction of the lineage tree of cardiac cells highlights the order of cell fate decisions in terms of cell type diversification (Figs 2,5) or spatial contribu-tions to the different cardiac compartments (Figs 3,5). Understanding how cells make such decisions, by balanc-ing pluripotent and differentiation factors or by receiving patterning cues, is important. Etv2, which encodes an ETS-domain transcription factor expressed in endothe-lial cells, is essential for endocardium development and is regulated by NKX2-5 (reF.173). Genetic-tracing experi-ments showed that Etv2-expressing cells give rise only to endocardial cells in the heart. However, in the absence of Etv2, the cells that should have expressed this tran-scription factor differentiate into other muscle lineages, including the myocardium174.
At E8.5–E10.5, forced Vegf expression favours the differentiation of endocardial cells at the expense of myocardial cells in the outflow tract, suggesting that VEGF has a role in the cell fate decision of bipotent, Isl1-expressing progenitor cells175. Notch signalling has been identified as instructive for the differentiation of contractile versus conductive cardiomyocytes. Induced expression of the Notch intracellular domain in the myo-cardium changed the cell phenotype so that it became indistinguishable from Purkinje cells, as evidenced by upregulation of Cntn2 and Cx40 (also known as Gja5) expression and by the characteristic action potential patterns176. In the epicardium, smooth muscle or fibro-blast progenitor cells are established before EMT and delamination into the ventricular walls177,178. Deletion of either Pdgfra or Pdgfrb results in a deficit in fibroblast
and smooth muscle progenitor cells, respectively, which do not undergo EMT177,178. Furthermore, the expression of the epicardial transcription factor TCF21, which is required for fibroblast EMT, becomes restricted to this cell type before delamination144. In the absence of TCF21, fibroblast progenitor cells in the epicardium activate Acta2 expression179, indicating cell fate plasticity.
Single- cell analysis of cardiogenesisRegional heterogeneity is observed at the cell population level in cardiac progenitors (for example, in the SHF), but these observations do not extend to potential het-erogeneity between cells within a region, which might influence their behaviour and future integration into the heart and might also reflect different states in cell cycle or cell lineage progression. This level of resolution requires single- cell analyses. We discuss data generated in the past 3 years by single- cell analysis of embryonic cardiac material.
Heart development. Single- cell RNA sequencing has been reported for cells isolated from different regions of the developing heart from E8.5 to postnatal day 21 (reFs180,181). Novel markers, as well as those already known, were identified for each region and for differ-ent cell types at different stages of development. These data provide a spatiotemporal resource of markers and potential regulators for normal cardiogenesis and open the way for similar studies of mutant hearts, which has relevance to congenital heart disease. Single- cell data also provide insight into transcriptional heterogen-eity within a particular region at a given time point. Whereas cell proliferation versus non- proliferation does not result in widely different transcriptomes, cell subpopulations with more heterogeneity are detected by these analyses (for example, ventricular cardiomyocytes expressing fibroblast- type extracellular matrix genes). Developmental profiles of individual cardiomyocytes show some overlap between developmental stages, indi-cating that maturation does not progress with a high degree of synchrony. Rare, consistently immature cells at later developmental stages were detected. Whether these cells have more regenerative potential than the more mature cardiomyocytes remains to be investigated.
Developmental trajectories of progenitor cells. In another study, cells from embryonic hearts at E8.5 and E9.5 or whole embryos at E7.5 were sorted using flu-orescent markers under the control of Isl1 or Nkx2-5 regulatory sequences before single- cell RNA sequenc-ing to establish their transcriptome profile or assay for transposase- accessible chromatin (ATAC) sequencing to characterize their chromatin landscape182. Using bioinformatic tools, developmental trajectories of the cells expressing Nkx2-5 or Isl1 were proposed, with cell subpopulations identified on the basis of tran-scriptome cluster analysis. For example, one Nkx2-5 cluster mainly represented cells present at E7.5 before the heart tube forms. Novel observations that emerged from this analysis include transcripts of posterior Hox genes present in a specific Isl1 cluster, whereas anterior Hox genes known to be expressed in the SHF showed
www.nature.com/nrcardio
R e v i e w s
cross- cluster distribution. In general, Nkx2-5 clusters were associated with a trajectory towards cardiomyo-cyte differentiation, whereas Isl1 clusters marked two trajectories towards myocardial and endothelial fates, with evidence for an intermediate cell state. Pseudo- time reconstructions of trajectories indicated transcriptome heterogeneity among cells collected at the same time point, consistent with non- synchronous maturation. Manipulation of Nkx2-5 confirmed that Nkx2-5 expres-sion promotes a cardiomyocyte trajectory, whereas when Isl1 was mutated, cells were stalled in a transition state and did not progress towards endothelial or cardiomyo-cyte fates, possibly owing to a reduction in cell cycling. The results of single- cell ATAC sequencing document dynamic changes in chromatin accessibility in different cell subpopulations. The absence of ISL1 led to a more open chromatin configuration, suggesting that this is a feature of the transition state. In conjunction with the transcriptome data, single- cell ATAC sequencing gives new insight into the transcription factors that orches-trate early heart formation. Thus, this approach, based on bioinformatic reconstructions, provides interesting material for study. However, the trajectories remain to be validated in vivo. Use of the whole embryo for the E7.5 time point potentially complicates the analysis, because both Isl1 and Nkx2-5 are also expressed in non- cardiac cell types, such as in the neural tube or gut endoderm.
Single- cell profiling of early cardiac progenitors marked by Mesp1 expression. Transcriptome profiling on Mesp1–rtTA;tetO–H2B–GFP cell populations iso-lated by flow cytometry 6 h after doxycycline induction at E6.25 or E7.25 showed expression of many genes asso-ciated with Mesp1 regulation and function26 and of early markers of cardiovascular progenitors such as Kdr and Pdgfra119. Transcripts of genes associated with cardiovas-cular development, encoding both transcription factors and signalling pathway components, were also present. Expression of genes that mark the SHF, such as Foxc1, Foxc2 or Hoxb1, was higher in cells at E7.5. Single- cell PCR analysis detected transcripts of SHF genes in <10% of E6.5 cells, whereas this number increased tenfold at E7.5. Consistent with the presence of unipotential or bipotential cells in the first or second cardiac lineage, respectively, single cells at E6.5 mainly expressed either the myocardial Myl7 or endothelial Etv2 marker genes, whereas more cells co- expressing these markers were detected at E7.5 (reF.26).
A comprehensive, single- cell RNA sequencing analy-sis at E6.75 and E7.25 of fluorescent cells that had expressed Mesp1 provides further insights into cardiac progenitor cell programming183. Five different clusters were highlighted at E7.25, representing distinct cell lineage profiles consistent with cell fate diversification of Mesp1-expressing progenitors into endothelial and endocardial cells, cardiomyocytes and SHF progeni-tors, with one of the SHF clusters also expressing phar-yngeal mesoderm markers, including early markers of skeletal myogenesis, thus capturing this myocardial cell sublineage (Fig. 2). A fifth cluster seems not to repre-sent cardiovascular progenitors because this cell clus-ter expressed endoderm marker genes such as Sox17
or Foxa2. However, in view of the finding that Foxa2 is also expressed in early cardiovascular progenitors42, this interpretation merits further investigation. This analy-sis would suggest that cardiac progenitors segregate early into distinct cardiovascular cell lineages, as indi-cated by the heterogeneity of transcriptome profiles of Mesp1-expressing cells at E7.25. The analysis also reveals novel markers of these early cell types that can now be exploited. Again, this type of reconstruction provides interesting material for study, but it should be noted that the cells selected for cluster analysis represented 10% of the total number of cells analysed. The other cells might reflect more of a continuum or more variation of desti-nation cell types. In vivo validation of the proposed cell lineage trajectories, in terms of transcriptomic changes, is currently lacking.
Single- cell profiling of early cardiac progenitors in the absence of Mesp1 expression. Both MESP1 and MESP2 are expressed in early cardiac progenitors184 and regulate EMT when the cardiac progenitors delaminate from the primitive streak185,186. MESP1 controls the directed migration of these cells, with Prickle1 and Rasgrp3 identi-fied as MESP1-specific target genes185. Experiments with embryonic stem cells also suggest that both genes regu-late cardiovascular progenitor cell specification185,187–189. Mesp2-mutant mice do not have heart defects; in Mesp1-mutant mice, partial compensation by Mesp2 leads to formation of hearts in which the defects reflect problems with cardiac progenitor cell migration24. In the absence of both Mesp1 and Mesp2, the mesoderm does not form at gastrulation184. Single- cell transcriptome analysis of cells isolated from Mesp1-mutant mice at E6.75 com-pared with wild- type mice has provided more insight into the role of MESP1 as a crucial regulator of the onset of cardiac mesoderm formation183 (Fig. 4c). In the absence of MESP1, transcripts for genes normally expressed in the embryo before gastrulation remain upregulated in cells that activated a Mesp1 reporter and thus normally are present in the primitive streak and nascent meso-derm. The upregulated genes include those that mark the epiblast such as Cdh1, and also genes associated with pluripotency such as Nanog or Pou5f1. Downregulated genes included those implicated in EMT (Snail), migra-tion (Rasgrp3) and cardiovascular commitment (such as Etv2, Kdr, Gata4, Hand1 and Pdgfra) as well as genes such as Myl7, indicating that cardiac priming extends to contractile- protein genes expressed later in the first differentiating cardiomyocytes.
The results of single- cell analysis discussed here rep-resent the first sets of data for cardiogenesis that have been generated by this approach, which is becoming widely used in lineage analyses. A major advantage of single- cell analysis is that the transcriptomes obtained are no longer an average of an heterogeneous cell popu-lation; therefore, the extent of heterogeneity can be quantified. This approach has a number of caveats, such as the level of technical and biological noise, which can be considerable so that it is important to perform appropriate controls and to analyse large numbers of cells. Another important limitation is the current lack of precise spatiotemporal information for individual cells.
R e v i e w s
Dissection of defined regions of the heart or of heart fields at specific times, together with cell sorting based on specific fluorescent markers, begins to provide some degree of spatiotemporal resolution. However, in the future, much more precise parameters will need to be used. Specific staging, based on anatomical criteria4,190, is required at early developmental stages, given the marked variations within a mouse litter. In terms of spatial resolution, isolation of cells from frozen sec-tions should provide more exact spatial parameters. We are still in the early days of single- cell analysis in cardiogenesis. Although this technique has provided novel insights into the role of genes such as Isl1 and Mesp1 in cardiac development, the contribution of single- cell analysis to understanding the deployment of cell lineages that form the heart remains at present limited. Defining cell clusters and drawing cell lineage trajectories computationally is always possible with the data. Previous knowledge on cell lineages is used to val-idate this novel technology. However, single- cell RNA sequencing data will have a major effect on our under-standing of cardiogenesis only when novel markers, cell populations or cell lineage trajectories are validated functionally in vivo.
ConclusionsConsiderable progress has been made in understand-ing cardiac progenitor cell specification and deploy-ment. A lineage tree of cardiac cells is emerging, which still requires clarifications (for example, at the top of the tree) for a more complete characterization of the progenitors in the primitive streak, extending beyond Mesp1 expression. How overlapping cell lineage
contributions provide robustness for heart development is just beginning to be understood. Many of the conclu-sions on cardiac gene regulatory networks are currently based on the analysis of populations of cells; therefore, much of the heterogeneity, which is beginning to be revealed by single- cell analysis, is averaged out. Single- cell analyses represent a major way forward towards understanding cell lineage progression. The challenges that face single- cell analyses are considerable in terms of linking the data to the spatiotemporal characteris-tics of individual cells. The molecular fine- tuning of cell lineage segregation and cardiac progenitor cell fate will require technical refinement to be able to equate these parameters to the spatiotemporal behaviour of the wild- type and mutant cells concerned. At present, the link is still indirect between the molecular charac-teristics, notably gene expression and gene regulatory circuitry, of cardiac progenitors and their cell lineage affiliation on the basis of clonal analyses. The potential of DNA bar coding, by the introduction, for example, of scars by random CRISPR–Cas9 genetic editing that marks single cells, combined with transcriptome and other types of analyses on the same cell from wild- type or mutant backgrounds191, opens up new perspectives for directly equating the two levels of observation. In addition to a better understanding of congenital heart malformations, the dissection of the molecular mech-anisms of cell lineage deployment will be important in the context of cardiac regenerative medicine, to engi-neer cardiac tissue with controlled properties or to stimulate repair processes in situ.
Published online xx xx xxxx
1. Galli, D. et al. Atrial myocardium derives from the posterior region of the second heart field, which acquires left- right identity as Pitx2c is expressed. Development 135, 1157–1167 (2008).
2. Kelly, R. G., Brown, N. A. & Buckingham, M. E. The arterial pole of the mouse heart forms from Fgf10-expressing cells in pharyngeal mesoderm. Dev. Cell 1, 435–440 (2001).
3. Zaffran, S., Kelly, R. G., Meilhac, S. M., Buckingham, M. E. & Brown, N. A. Right ventricular myocardium derives from the anterior heart field. Circ. Res. 95, 261–268 (2004).
4. Le Garrec, J. F. et al. A predictive model of asymmetric morphogenesis from 3D reconstructions of mouse heart looping dynamics. Elife 6, e28951 (2017).
5. Meilhac, S. M., Esner, M., Kerszberg, M., Moss, J. E. & Buckingham, M. E. Oriented clonal cell growth in the developing mouse myocardium underlies cardiac morphogenesis. J. Cell Biol. 164, 97–109 (2004).
6. Christoffels, V. M. et al. Chamber formation and morphogenesis in the developing mammalian heart. Dev. Biol. 223, 266–278 (2000).
7. Hoffman, J. I. & Kaplan, S. The incidence of congenital heart disease. J. Am. Coll. Cardiol. 39, 1890–1900 (2002).
8. de la Cruz, M., Sanchez- Gomez, C. & Palomino, M. The primitive cardiac regions in the straight heart tube (stage 9-) and their anatomical expression in the mature heart: an experimental study in the chick embryo. J. Anat. 165, 121–131 (1989).
9. Garcia- Martinez, V. & Schoenwolf, G. C. Primitive- streak origin of the cardiovascular system in avian embryos. Dev. Biol. 159, 706–719 (1993).
10. Mikawa, T., Borisov, A., Brown, A. M. & Fischman, D. A. Clonal analysis of cardiac morphogenesis in the chicken embryo using a replication- defective retrovirus: I. Formation of the ventricular myocardium. Dev. Dyn. 193, 11–23 (1992).
11. Rawles, M. E. The heart- forming areas of the early chick blastoderm. Physiol. Zool. 16, 22–42 (1943).
12. Redkar, A., Montgomery, M. & Litvin, J. Fate map of early avian cardiac progenitor cells. Development 128, 2269–2279 (2001).
13. Stalsberg, H. The origin of heart asymmetry: right and left contributions to the early chick embryo heart. Dev. Biol. 19, 109–127 (1969).
14. Waldo, K. L., Lo, C. W. & Kirby, M. L. Connexin 43 expression reflects neural crest patterns during cardiovascular development. Dev. Biol. 208, 307–323 (1999).
15. Buckingham, M., Biben, C. & Lawson, K. in Genetic Control of Heart Development (eds Olson, E. N., Harvey, R. P., Schulz, R. A. & Altman, J. S.) 31–33 (Strasbourg : HFSP Pub, 1997).
16. Lawson, K. A. & Pedersen, R. A. Cell fate, morphogenetic movement and population kinetics of embryonic endoderm at the time of germ layer formation in the mouse. Development 101, 627–652 (1987).
17. Tzouanacou, E., Wegener, A., Wymeersch, F. J., Wilson, V. & Nicolas, J. F. Redefining the progression of lineage segregations during mammalian embryogenesis by clonal analysis. Dev. Cell 17, 365–376 (2009).
18. Kinder, S. J. et al. The orderly allocation of mesodermal cells to the extraembryonic structures and the anteroposterior axis during gastrulation of the mouse embryo. Development 126, 4691–4701 (1999).
19. Ivanovitch, K., Temino, S. & Torres, M. Live imaging of heart tube development in mouse reveals alternating phases of cardiac differentiation and morphogenesis. Elife 6, e30668 (2017).
20. Kaufman, M. H. & Navaratnam, V. Early differentiation of the heart in mouse embryos. J. Anat. 133, 235–246 (1981).
21. Tam, P. P., Parameswaran, M., Kinder, S. J. & Weinberger, R. P. The allocation of epiblast cells to the embryonic heart and other mesodermal lineages: the role of ingression and tissue movement during gastrulation. Development 124, 1631–1642 (1997).
22. Meilhac, S. M., Esner, M., Kelly, R. G., Nicolas, J. F. & Buckingham, M. E. The clonal origin of myocardial cells in different regions of the embryonic mouse heart. Dev. Cell 6, 685–698 (2004).
23. Ragni, C. V. et al. Amotl1 mediates sequestration of the Hippo effector Yap1 downstream of Fat4 to restrict heart growth. Nat. Commun. 8, 14582 (2017).
24. Saga, Y. et al. MesP1 is expressed in the heart precursor cells and required for the formation of a single heart tube. Development 126, 3437–3447 (1999).
25. Buckingham, M. E. & Meilhac, S. M. Tracing cells for tracking cell lineage and clonal behavior. Dev. Cell 21, 394–409 (2011).
26. Lescroart, F. et al. Early lineage restriction in temporally distinct populations of Mesp1 progenitors during mammalian heart development. Nat. Cell Biol. 16, 829–840 (2014).
27. Chabab, S. et al. Uncovering the number and clonal dynamics of Mesp1 progenitors during heart morphogenesis. Cell Rep. 14, 1–10 (2016).
28. Zong, H., Espinosa, J. S., Su, H. H., Muzumdar, M. D. & Luo, L. Mosaic analysis with double markers in mice. Cell 121, 479–492 (2005).
29. Devine, W. P., Wythe, J. D., George, M., Koshiba- Takeuchi, K. & Bruneau, B. G. Early patterning and specification of cardiac progenitors in gastrulating mesoderm. Elife 3, e03848 (2014).
30. Franco, D. et al. Left and right ventricular contributions to the formation of the interventricular septum in the mouse heart. Dev. Biol. 294, 366–375 (2006).
31. Buckingham, M. Gene regulatory networks and cell lineages that underlie the formation of skeletal muscle. Proc. Natl Acad. Sci. USA 114, 5830–5837 (2017).
32. Lescroart, F. et al. Clonal analysis reveals common lineage relationships between head muscles and second heart field derivatives in the mouse embryo. Development 137, 3269–3279 (2010).
33. Lescroart, F. et al. Clonal analysis reveals a common origin between nonsomite- derived neck muscles and
Nature reviews | Cardiology
www.nature.com/nrcardio
R e v i e w s
heart myocardium. Proc. Natl Acad. Sci. USA 112, 1446–1451 (2015).
34. Lescroart, F., Mohun, T., Meilhac, S. M., Bennett, M. & Buckingham, M. Lineage tree for the venous pole of the heart: clonal analysis clarifies controversial genealogy based on genetic tracing. Circ. Res. 111, 1313–1322 (2012).
35. Dominguez, J. N., Meilhac, S. M., Bland, Y. S., Buckingham, M. E. & Brown, N. A. Asymmetric fate of the posterior part of the second heart field results in unexpected left/right contributions to both poles of the heart. Circ. Res. 111, 1323–1335 (2012).
36. Bajolle, F. et al. Conotruncal defects associated with anomalous pulmonary venous connections. Arch. Cardiovasc. Dis. 102, 105–110 (2009).
37. Meilhac, S. M. et al. A retrospective clonal analysis of the myocardium reveals two phases of clonal growth in the developing mouse heart. Development 130, 3877–3889 (2003).
38. Del Monte- Nieto, G. et al. Control of cardiac jelly dynamics by NOTCH1 and NRG1 defines the building plan for trabeculation. Nature 557, 439–445 (2018).
39. Tian, X. et al. Identification of a hybrid myocardial zone in the mammalian heart after birth. Nat. Commun. 8, 87 (2017).
40. Miquerol, L. et al. Resolving cell lineage contributions to the ventricular conduction system with a Cx40-GFP allele: a dual contribution of the first and second heart fields. Dev. Dyn. 242, 665–677 (2013).
41. Miquerol, L. et al. Biphasic development of the mammalian ventricular conduction system. Circ. Res. 107, 153–161 (2010).
42. Bardot, E. et al. Foxa2 identifies a cardiac progenitor population with ventricular differentiation potential. Nat. Commun. 8, 14428 (2017).
43. Ang, S. L. et al. The formation and maintenance of the definitive endoderm lineage in the mouse: involvement of HNF3/forkhead proteins. Development 119, 1301–1315 (1993).
44. Kinder, S. J. et al. The organizer of the mouse gastrula is composed of a dynamic population of progenitor cells for the axial mesoderm. Development 128, 3623–3634 (2001).
45. Park, E. J. et al. System for tamoxifen- inducible expression of cre- recombinase from the Foxa2 locus in mice. Dev. Dyn. 237, 447–453 (2008).
46. Verzi, M. P., McCulley, D. J., De Val, S., Dodou, E. & Black, B. L. The right ventricle, outflow tract, and ventricular septum comprise a restricted expression domain within the secondary/anterior heart field. Dev. Biol. 287, 134–145 (2005).
47. Buckingham, M., Meilhac, S. & Zaffran, S. Building the mammalian heart from two sources of myocardial cells. Nat. Rev. Genet. 6, 826–835 (2005).
48. Bruneau, B. G. et al. Chamber- specific cardiac expression of Tbx5 and heart defects in Holt- Oram syndrome. Dev. Biol. 211, 100–108 (1999).
49. Rana, M. S. et al. Tbx1 coordinates addition of posterior second heart field progenitor cells to the arterial and venous poles of the heart. Circ. Res. 115, 790–799 (2014).
50. Vincentz, J. W., Toolan, K. P., Zhang, W. & Firulli, A. B. Hand factor ablation causes defective left ventricular chamber development and compromised adult cardiac function. PLOS Genet. 13, e1006922 (2017).
51. Barnes, R. M., Firulli, B. A., Conway, S. J., Vincentz, J. W. & Firulli, A. B. Analysis of the Hand1 cell lineage reveals novel contributions to cardiovascular, neural crest, extra- embryonic, and lateral mesoderm derivatives. Dev. Dyn. 239, 3086–3097 (2010).
52. Liang, X. et al. HCN4 dynamically marks the first heart field and conduction system precursors. Circ. Res. 113, 399–407 (2013).
53. Spater, D. et al. A HCN4+ cardiomyogenic progenitor derived from the first heart field and human pluripotent stem cells. Nat. Cell Biol. 15, 1098–1106 (2013).
54. Fujii, M. et al. Sfrp5 identifies murine cardiac progenitors for all myocardial structures except for the right ventricle. Nat. Commun. 8, 14664 (2017).
55. Watanabe, Y. et al. Role of mesodermal FGF8 and FGF10 overlaps in the development of the arterial pole of the heart and pharyngeal arch arteries. Circ. Res. 106, 495–503 (2010).
56. Cai, C. L. et al. Isl1 identifies a cardiac progenitor population that proliferates prior to differentiation and contributes a majority of cells to the heart. Dev. Cell 5, 877–889 (2003).
57. Engleka, K. A. et al. Islet1 derivatives in the heart are of both neural crest and second heart field origin. Circ. Res. 110, 922–926 (2012).
58. Ma, Q., Zhou, B. & Pu, W. T. Reassessment of Isl1 and Nkx2-5 cardiac fate maps using a Gata4-based reporter of Cre activity. Dev. Biol. 323, 98–104 (2008).
59. Sun, Y. et al. Islet 1 is expressed in distinct cardiovascular lineages, including pacemaker and coronary vascular cells. Dev. Biol. 304, 286–296 (2007).
60. Prall, O. W. et al. An Nkx2-5/Bmp2/Smad1 negative feedback loop controls heart progenitor specification and proliferation. Cell 128, 947–959 (2007).
61. Stanley, E. G. et al. Efficient Cre- mediated deletion in cardiac progenitor cells conferred by a 3’UTR- ires-Cre allele of the homeobox gene Nkx2-5. Int. J. Dev. Biol. 46, 431–439 (2002).
62. Seo, S. & Kume, T. Forkhead transcription factors, Foxc1 and Foxc2, are required for the morphogenesis of the cardiac outflow tract. Dev. Biol. 296, 421–436 (2006).
63. Zhou, Z. et al. Temporally distinct Six2-positive second heart field progenitors regulate mammalian heart development and disease. Cell Rep. 18, 1019–1032 (2017).
64. Baldini, A., Fulcoli, F. G. & Illingworth, E. Tbx1: transcriptional and developmental functions. Curr. Top. Dev. Biol. 122, 223–243 (2017).
65. Theveniau- Ruissy, M. et al. The del22q11.2 candidate gene Tbx1 controls regional outflow tract identity and coronary artery patterning. Circ. Res. 103, 142–148 (2008).
66. Huynh, T., Chen, L., Terrell, P. & Baldini, A. A fate map of Tbx1 expressing cells reveals heterogeneity in the second cardiac field. Genesis 45, 470–475 (2007).
67. Bajolle, F. et al. Myocardium at the base of the aorta and pulmonary trunk is prefigured in the outflow tract of the heart and in subdomains of the second heart field. Dev. Biol. 313, 25–34 (2008).
68. Kelly, R. G. & Papaioannou, V. E. Visualization of outflow tract development in the absence of Tbx1 using an FgF10 enhancer trap transgene. Dev. Dyn. 236, 821–828 (2007).
69. Cortes, C., Francou, A., De Bono, C. & Kelly, R. G. Epithelial properties of the second heart field. Circ. Res. 122, 142–154 (2018).
70. Zhou, L. et al. Tbx5 and Osr1 interact to regulate posterior second heart field cell cycle progression for cardiac septation. J. Mol. Cell Cardiol. 85, 1–12 (2015).
71. Tian, Y. et al. Characterization and in vivo pharmacological rescue of a Wnt2-Gata6 pathway required for cardiac inflow tract development. Dev. Cell 18, 275–287 (2010).
72. Christoffels, V. M. et al. Formation of the venous pole of the heart from an Nkx2-5-negative precursor population requires Tbx18. Circ. Res. 98, 1555–1563 (2006).
73. Mommersteeg, M. T. et al. The sinus venosus progenitors separate and diversify from the first and second heart fields early in development. Cardiovasc. Res. 87, 92–101 (2010).
74. Bertrand, N. et al. Hox genes define distinct progenitor sub- domains within the second heart field. Dev. Biol. 353, 266–274 (2011).
75. Furtado, M. B., Biben, C., Shiratori, H., Hamada, H. & Harvey, R. P. Characterization of Pitx2c expression in the mouse heart using a reporter transgene. Dev. Dyn. 240, 195–203 (2011).
76. Bajolle, F. et al. Rotation of the myocardial wall of the outflow tract is implicated in the normal positioning of the great arteries. Circ. Res. 98, 421–428 (2006).
77. Liu, C. et al. Pitx2c patterns anterior myocardium and aortic arch vessels and is required for local cell movement into atrioventricular cushions. Development 129, 5081–5091 (2002).
78. Dyer, L. A. & Kirby, M. L. The role of secondary heart field in cardiac development. Dev. Biol. 336, 137–144 (2009).
79. Kelly, R. G., Buckingham, M. E. & Moorman, A. F. Heart fields and cardiac morphogenesis. Cold Spring Harb. Perspect. Med. 4, a015750 (2014).
80. Vincent, S. D. & Buckingham, M. E. How to make a heart: the origin and regulation of cardiac progenitor cells. Curr. Top. Dev. Biol. 90, 1–41 (2010).
81. Jain, R. et al. HEART DEVELOPMENT. Integration of Bmp and Wnt signaling by Hopx specifies commitment of cardiomyoblasts. Science 348, aaa6071(2015).
82. Kang, J., Nathan, E., Xu, S. M., Tzahor, E. & Black, B. L. Isl1 is a direct transcriptional target of Forkhead transcription factors in second- heart-field- derived mesoderm. Dev. Biol. 334, 513–522 (2009).
83. Kappen, C. & Salbaum, J. M. Identification of regulatory elements in the Isl1 gene locus. Int. J. Dev. Biol. 53, 935–946 (2009).
84. Dodou, E., Verzi, M. P., Anderson, J. P., Xu, S. M. & Black, B. L. Mef2c is a direct transcriptional target of ISL1 and GATA factors in the anterior heart field during mouse embryonic development. Development 131, 3931–3942 (2004).
85. Watanabe, Y. et al. Fibroblast growth factor 10 gene regulation in the second heart field by Tbx1, Nkx2-5, and Islet1 reveals a genetic switch for down- regulation in the myocardium. Proc. Natl Acad. Sci. USA 109, 18273–18280 (2012).
86. Guo, C. et al. A Tbx1-Six1/Eya1-Fgf8 genetic pathway controls mammalian cardiovascular and craniofacial morphogenesis. J. Clin. Invest. 121, 1585–1595 (2011).
87. Yuan, X. et al. Disruption of spatiotemporal hypoxic signaling causes congenital heart disease in mice. J. Clin. Invest. 127, 2235–2248 (2017).
88. Liberatore, C. M., Searcy- Schrick, R. D., Vincent, E. B. & Yutzey, K. E. Nkx-2.5 gene induction in mice is mediated by a Smad consensus regulatory region. Dev. Biol. 244, 243–256 (2002).
89. Lien, C. L., McAnally, J., Richardson, J. A. & Olson, E. N. Cardiac- specific activity of an Nkx2-5 enhancer requires an evolutionarily conserved Smad binding site. Dev. Biol. 244, 257–266 (2002).
90. von Both, I. et al. Foxh1 is essential for development of the anterior heart field. Dev. Cell 7, 331–345 (2004).
91. Dorn, T. et al. Direct nkx2-5 transcriptional repression of isl1 controls cardiomyocyte subtype identity. Stem Cells 33, 1113–1129 (2015).
92. Xie, L. et al. Tbx5-hedgehog molecular networks are essential in the second heart field for atrial septation. Dev. Cell 23, 280–291 (2012).
93. Snarr, B. S., Wirrig, E. E., Phelps, A. L., Trusk, T. C. & Wessels, A. A spatiotemporal evaluation of the contribution of the dorsal mesenchymal protrusion to cardiac development. Dev. Dyn. 236, 1287–1294 (2007).
94. Snarr, B. S. et al. Isl1 expression at the venous pole identifies a novel role for the second heart field in cardiac development. Circ. Res. 101, 971–974 (2007).
95. Goddeeris, M. M. et al. Intracardiac septation requires hedgehog- dependent cellular contributions from outside the heart. Development 135, 1887–1895 (2008).
96. Hoffmann, A. D., Peterson, M. A., Friedland- Little, J. M., Anderson, S. A. & Moskowitz, I. P. Sonic hedgehog is required in pulmonary endoderm for atrial septation. Development 136, 1761–1770 (2009).
97. Hoffmann, A. D. et al. Foxf genes integrate tbx5 and hedgehog pathways in the second heart field for cardiac septation. PLOS Genet. 10, e1004604 (2014).
98. Zhou, L. et al. Gata4 potentiates second heart field proliferation and Hedgehog signaling for cardiac septation. Proc. Natl Acad. Sci. USA 114, E1422–E1431 (2017).
99. Durocher, D., Charron, F., Warren, R., Schwartz, R. J. & Nemer, M. The cardiac transcription factors Nkx2-5 and GATA-4 are mutual cofactors. EMBO J. 16, 5687–5696 (1997).
100. Garg, V. et al. GATA4 mutations cause human congenital heart defects and reveal an interaction with TBX5. Nature 424, 443–447 (2003).
101. Maitra, M. et al. Interaction of Gata4 and Gata6 with Tbx5 is critical for normal cardiac development. Dev. Biol. 326, 368–377 (2009).
102. Ang, Y. S. et al. Disease model of GATA4 mutation reveals transcription factor cooperativity in human cardiogenesis. Cell 167, 1734–1749.e22 (2016).
103. Luna- Zurita, L. et al. Complex interdependence regulates heterotypic transcription factor distribution and coordinates cardiogenesis. Cell 164, 999–1014 (2016).
104. Bouveret, R. et al. NKX2-5 mutations causative for congenital heart disease retain functionality and are directed to hundreds of targets. Elife 4, e06942 (2015).
105. Witzel, H. R. et al. The LIM protein Ajuba restricts the second heart field progenitor pool by regulating Isl1 activity. Dev. Cell 23, 58–70 (2012).
106. Caputo, L. et al. The Isl1/Ldb1 complex orchestrates genome- wide chromatin organization to instruct differentiation of multipotent cardiac progenitors. Cell Stem Cell 17, 287–299 (2015).
107. Stoller, J. Z. et al. Ash2l interacts with Tbx1 and is required during early embryogenesis. Exp. Biol. Med. (Maywood) 235, 569–576 (2010).
Nature reviews | Cardiology
R e v i e w s
108. Fulcoli, F. G. et al. Rebalancing gene haploinsufficiency in vivo by targeting chromatin. Nat. Commun. 7, 11688 (2016).
109. Chen, L. et al. Transcriptional control in cardiac progenitors: Tbx1 interacts with the BAF chromatin remodeling complex and regulates Wnt5a. PLOS Genet. 8, e1002571 (2012).
110. Takeuchi, J. K. & Bruneau, B. G. Directed transdifferentiation of mouse mesoderm to heart tissue by defined factors. Nature 459, 708–711 (2009).
111. Poleshko, A. et al. Genome- nuclear lamina interactions regulate cardiac stem cell lineage restriction. Cell 171, 573–587.e14 (2017).
112. Klattenhoff, C. A. et al. Braveheart, a long noncoding RNA required for cardiovascular lineage commitment. Cell 152, 570–583 (2013).
113. Xue, Z. et al. A G- rich motif in the lncRNA braveheart interacts with a zinc- finger transcription factor to specify the cardiovascular lineage. Mol. Cell 64, 37–50 (2016).
114. Grote, P. et al. The tissue- specific lncRNA Fendrr is an essential regulator of heart and body wall development in the mouse. Dev. Cell 24, 206–214 (2013).
115. Epstein, J. A. et al. Migration of cardiac neural crest cells in Splotch embryos. Development 127, 1869–1878 (2000).
116. Jiang, X., Rowitch, D. H., Soriano, P., McMahon, A. P. & Sucov, H. M. Fate of the mammalian cardiac neural crest. Development 127, 1607–1616 (2000).
117. Kimura, K., Ieda, M. & Fukuda, K. Development, maturation, and transdifferentiation of cardiac sympathetic nerves. Circ. Res. 110, 325–336 (2012).
118. Saga, Y., Kitajima, S. & Miyagawa- Tomita, S. Mesp1 expression is the earliest sign of cardiovascular development. Trends Cardiovasc. Med. 10, 345–352 (2000).
119. Kattman, S. J., Huber, T. L. & Keller, G. M. Multipotent flk-1+ cardiovascular progenitor cells give rise to the cardiomyocyte, endothelial, and vascular smooth muscle lineages. Dev. Cell 11, 723–732 (2006).
120. Moretti, A. et al. Multipotent embryonic isl1+ progenitor cells lead to cardiac, smooth muscle, and endothelial cell diversification. Cell 127, 1151–1165 (2006).
121. Wu, S. M. et al. Developmental origin of a bipotential myocardial and smooth muscle cell precursor in the mammalian heart. Cell 127, 1137–1150 (2006).
122. Nakano, A., Nakano, H., Smith, K. A. & Palpant, N. J. The developmental origins and lineage contributions of endocardial endothelium. Biochim. Biophys. Acta 1863, 1937–1947 (2016).
123. Ema, M., Takahashi, S. & Rossant, J. Deletion of the selection cassette, but not cis- acting elements, in targeted Flk1-lacZ allele reveals Flk1 expression in multipotent mesodermal progenitors. Blood 107, 111–117 (2006).
124. Motoike, T., Markham, D. W., Rossant, J. & Sato, T. N. Evidence for novel fate of Flk1+ progenitor: contribution to muscle lineage. Genesis 35, 153–159 (2003).
125. de la Pompa, J. L. et al. Role of the NF- ATc transcription factor in morphogenesis of cardiac valves and septum. Nature 392, 182–186 (1998).
126. Misfeldt, A. M. et al. Endocardial cells are a distinct endothelial lineage derived from Flk1+ multipotent cardiovascular progenitors. Dev. Biol. 333, 78–89 (2009).
127. Wu, B. et al. Nfatc1 coordinates valve endocardial cell lineage development required for heart valve formation. Circ. Res. 109, 183–192 (2011).
128. Nakano, H. et al. Haemogenic endocardium contributes to transient definitive haematopoiesis. Nat. Commun. 4, 1564 (2013).
129. Peterkin, T., Gibson, A. & Patient, R. Common genetic control of haemangioblast and cardiac development in zebrafish. Development 136, 1465–1474 (2009).
130. Schulte, I., Schlueter, J., Abu- Issa, R., Brand, T. & Manner, J. Morphological and molecular left- right asymmetries in the development of the proepicardium: a comparative analysis on mouse and chick embryos. Dev. Dyn. 236, 684–695 (2007).
131. Wessels, A. et al. Epicardially derived fibroblasts preferentially contribute to the parietal leaflets of the atrioventricular valves in the murine heart. Dev. Biol. 366, 111–124 (2012).
132. Zhou, B., von Gise, A., Ma, Q., Rivera- Feliciano, J. & Pu, W. T. Nkx2-5- and Isl1-expressing cardiac progenitors contribute to proepicardium. Biochem. Biophys. Res. Commun. 375, 450–453 (2008).
133. Cai, C. L. et al. A myocardial lineage derives from Tbx18 epicardial cells. Nature 454, 104–108 (2008).
134. Zhou, B. et al. Epicardial progenitors contribute to the cardiomyocyte lineage in the developing heart. Nature 454, 109–113 (2008).
135. Christoffels, V. M. et al. Tbx18 and the fate of epicardial progenitors. Nature 458, E9–E10 (2009).
136. Rudat, C. & Kispert, A. Wt1 and epicardial fate mapping. Circ. Res. 111, 165–169 (2012).
137. Martinez- Estrada, O. M. et al. Wt1 is required for cardiovascular progenitor cell formation through transcriptional control of Snail and E- cadherin. Nat. Genet. 42, 89–93 (2010).
138. Villa Del Campo, C. et al. Myc overexpression enhances of epicardial contribution to the developing heart and promotes extensive expansion of the cardiomyocyte population. Sci. Rep. 6, 35366 (2016).
139. Ivey, M. J. & Tallquist, M. D. Defining the cardiac fibroblast. Circ. J. 80, 2269–2276 (2016).
140. Kolditz, D. P. et al. Epicardium- derived cells in development of annulus fibrosis and persistence of accessory pathways. Circulation 117, 1508–1517 (2008).
141. Zhou, B., von Gise, A., Ma, Q., Hu, Y. W. & Pu, W. T. Genetic fate mapping demonstrates contribution of epicardium- derived cells to the annulus fibrosis of the mammalian heart. Dev. Biol. 338, 251–261 (2010).
142. Ali, S. R. et al. Developmental heterogeneity of cardiac fibroblasts does not predict pathological proliferation and activation. Circ. Res. 115, 625–635 (2014).
143. Moore- Morris, T. et al. Resident fibroblast lineages mediate pressure overload- induced cardiac fibrosis. J. Clin. Invest. 124, 2921–2934 (2014).
144. Acharya, A. et al. The bHLH transcription factor Tcf21 is required for lineage- specific EMT of cardiac fibroblast progenitors. Development 139, 2139–2149 (2012).
145. Kanisicak, O. et al. Genetic lineage tracing defines myofibroblast origin and function in the injured heart. Nat. Commun. 7, 12260 (2016).
146. Lie- Venema, H. et al. Origin, fate, and function of epicardium- derived cells (EPDCs) in normal and abnormal cardiac development. ScientificWorldJournal 7, 1777–1798 (2007).
147. Chen, Q. et al. Endothelial cells are progenitors of cardiac pericytes and vascular smooth muscle cells. Nat. Commun. 7, 12422 (2016).
148. Mikawa, T. & Fischman, D. A. Retroviral analysis of cardiac morphogenesis: discontinuous formation of coronary vessels. Proc. Natl Acad. Sci. USA 89, 9504–9508 (1992).
149. Sharma, B., Chang, A. & Red- Horse, K. Coronary artery development: progenitor cells and differentiation pathways. Annu. Rev. Physiol. 79, 1–19 (2017).
150. Tian, X., Pu, W. T. & Zhou, B. Cellular origin and developmental program of coronary angiogenesis. Circ. Res. 116, 515–530 (2015).
151. Red- Horse, K., Ueno, H., Weissman, I. L. & Krasnow, M. A. Coronary arteries form by developmental reprogramming of venous cells. Nature 464, 549–553 (2010).
152. Chen, H. I. et al. The sinus venosus contributes to coronary vasculature through VEGFC- stimulated angiogenesis. Development 141, 4500–4512 (2014).
153. Katz, T. C. et al. Distinct compartments of the proepicardial organ give rise to coronary vascular endothelial cells. Dev. Cell 22, 639–650 (2012).
154. Lavine, K. J. et al. Fibroblast growth factor signals regulate a wave of Hedgehog activation that is essential for coronary vascular development. Genes Dev. 20, 1651–1666 (2006).
155. Tian, X. et al. Subepicardial endothelial cells invade the embryonic ventricle wall to form coronary arteries. Cell Res. 23, 1075–1090 (2013).
156. Sharma, B. et al. Alternative progenitor cells compensate to rebuild the coronary vasculature in Elabela- and Apj- deficient hearts. Dev. Cell 42, 655–666.e3 (2017).
157. Wu, B. et al. Endocardial cells form the coronary arteries by angiogenesis through myocardial- endocardial VEGF signaling. Cell 151, 1083–1096 (2012).
158. Zhang, Z. & Zhou, B. Accelerated coronary angiogenesis by vegfr1-knockout endocardial cells. PLOS One 8, e70570 (2013).
159. Zhang, H. et al. Endocardium minimally contributes to coronary endothelium in the embryonic ventricular free walls. Circ. Res. 118, 1880–1893 (2016).
160. Cano, E. et al. Extracardiac septum transversum/proepicardial endothelial cells pattern embryonic coronary arterio- venous connections. Proc. Natl Acad. Sci. USA 113, 656–661 (2016).
161. Klotz, L. et al. Cardiac lymphatics are heterogeneous in origin and respond to injury. Nature 522, 62–67 (2015).
162. Ulvmar, M. H., Martinez- Corral, I., Stanczuk, L. & Makinen, T. Pdgfrb- Cre targets lymphatic endothelial cells of both venous and non- venous origins. Genesis 54, 350–358 (2016).
163. Bernanke, D. H. & Markwald, R. R. Migratory behavior of cardiac cushion tissue cells in a collagen- lattice culture system. Dev. Biol. 91, 235–245 (1982).
164. Markwald, R. R., Fitzharris, T. P. & Manasek, F. J. Structural development of endocardial cushions. Am. J. Anat. 148, 85–119 (1977).
165. Mjaatvedt, C. H., Lepera, R. C. & Markwald, R. R. Myocardial specificity for initiating endothelial- mesenchymal cell transition in embryonic chick heart correlates with a particulate distribution of fibronectin. Dev. Biol. 119, 59–67 (1987).
166. Ma, L., Lu, M. F., Schwartz, R. J. & Martin, J. F. Bmp2 is essential for cardiac cushion epithelial- mesenchymal transition and myocardial patterning. Development 132, 5601–5611 (2005).
167. de Lange, F. J. et al. Lineage and morphogenetic analysis of the cardiac valves. Circ. Res. 95, 645–654 (2004).
168. Waldo, K. et al. A novel role for cardiac neural crest in heart development. J. Clin. Invest. 103, 1499–1507 (1999).
169. Odelin, G. et al. Krox20 defines a subpopulation of cardiac neural crest cells contributing to arterial valves and bicuspid aortic valve. Development 145, https://doi.org/10.1242/dev.151944 (2018).
170. Lincoln, J., Alfieri, C. M. & Yutzey, K. E. Development of heart valve leaflets and supporting apparatus in chicken and mouse embryos. Dev. Dyn. 230, 239–250 (2004).
171. Eley, L. et al. A novel source of arterial valve cells linked to bicuspid aortic valve without raphe in mice. Elife 7, e34110 (2018).
172. Mifflin, J. J., Dupuis, L. E., Alcala, N. E., Russell, L. G. & Kern, C. B. Intercalated cushion cells within the cardiac outflow tract are derived from the myocardial troponin T type 2 (Tnnt2) Cre lineage. Dev. Dyn. 247, 1005–1017 (2018).
173. Ferdous, A. et al. Nkx2-5 transactivates the Ets- related protein 71 gene and specifies an endothelial/endocardial fate in the developing embryo. Proc. Natl Acad. Sci. USA 106, 814–819 (2009).
174. Rasmussen, T. L. et al. ER71 directs mesodermal fate decisions during embryogenesis. Development 138, 4801–4812 (2011).
175. He, Z., Grunewald, M., Dor, Y. & Keshet, E. VEGF regulates relative allocation of Isl1(+) cardiac progenitors to myocardial and endocardial lineages. Mech. Dev. 142, 40–49 (2016).
176. Rentschler, S. et al. Myocardial Notch signaling reprograms cardiomyocytes to a conduction- like phenotype. Circulation 126, 1058–1066 (2012).
177. Mellgren, A. M. et al. Platelet- derived growth factor receptor beta signaling is required for efficient epicardial cell migration and development of two distinct coronary vascular smooth muscle cell populations. Circ. Res. 103, 1393–1401 (2008).
178. Smith, C. L., Baek, S. T., Sung, C. Y. & Tallquist, M. D. Epicardial- derived cell epithelial- to-mesenchymal transition and fate specification require PDGF receptor signaling. Circ. Res. 108, e15–e26 (2011).
179. Braitsch, C. M., Combs, M. D., Quaggin, S. E. & Yutzey, K. E. Pod1/Tcf21 is regulated by retinoic acid signaling and inhibits differentiation of epicardium- derived cells into smooth muscle in the developing heart. Dev. Biol. 368, 345–357 (2012).
180. DeLaughter, D. M. et al. Single- cell resolution of temporal gene expression during heart development. Dev. Cell 39, 480–490 (2016).
181. Li, G. et al. Transcriptomic profiling maps anatomically patterned subpopulations among single embryonic cardiac cells. Dev. Cell 39, 491–507 (2016).
182. Jia, G. et al. Single- cell transcriptional regulations and accessible chromatin landscape of cell fate decisions in early heart development. bioRxiv https://doi.org/ 10.1101/210930 (2017).
183. Lescroart, F. et al. Defining the earliest step of cardiovascular lineage segregation by single- cell RNA- seq. Science 359, 1177–1181 (2018).
184. Kitajima, S., Takagi, A., Inoue, T. & Saga, Y. MesP1 and MesP2 are essential for the development of cardiac mesoderm. Development 127, 3215–3226 (2000).
185. Chiapparo, G. et al. Mesp1 controls the speed, polarity, and directionality of cardiovascular progenitor migration. J. Cell Biol. 213, 463–477 (2016).
www.nature.com/nrcardio
R e v i e w s
186. Lindsley, R. C. et al. Mesp1 coordinately regulates cardiovascular fate restriction and epithelial- mesenchymal transition in differentiating ESCs. Cell Stem Cell 3, 55–68 (2008).
187. Bondue, A. et al. Mesp1 acts as a master regulator of multipotent cardiovascular progenitor specification. Cell Stem Cell 3, 69–84 (2008).
188. Bondue, A. et al. Defining the earliest step of cardiovascular progenitor specification during embryonic stem cell differentiation. J. Cell Biol. 192, 751–765 (2011).
189. David, R. et al. MesP1 drives vertebrate cardiovascular differentiation through Dkk-1-mediated blockade of Wnt- signalling. Nat. Cell Biol. 10, 338–345 (2008).
190. Downs, K. M. & Davies, T. Staging of gastrulating mouse embryos by morphological landmarks in the
dissecting microscope. Development 118, 1255–1266 (1993).
191. Spanjaard, B. & Junker, J. P. Methods for lineage tracing on the organism- wide level. Curr. Opin. Cell Biol. 49, 16–21 (2017).
192. Meilhac, S. M., Lescroart, F., Blanpain, C. & Buckingham, M. E. Cardiac cell lineages that form the heart. Cold Spring Harb Perspect Med. 4, a013888 (2014).
193. Moore- Morris, T., Cattaneo, P., Puceat, M. & Evans, S. M. Origins of cardiac fibroblasts. J. Mol. Cell. Cardiol. 91, 1–5 (2016).
AcknowledgementsWork in the group of S.M.M. is supported by the Institut Imagine, the Institut Pasteur, INSERM and the Université Paris Descartes. M.E.B. acknowledges support from the
Institut Pasteur and the Centre National de la Recherche Scientifique (CNRS; UMR 3738). The authors thank R. Kelly (Institut de Biologie du Développement de Marseille, France) and S. Zaffran and F. Lescroart (Marseille Medical Genetics, France) for helpful comments on the manuscript.
Author contributionsBoth authors researched data for the article, discussed its content, wrote the manuscript and reviewed and edited it before submission.
Competing interestsThe authors declare no competing interests.
Publisher’s noteSpringer Nature remains neutral with regard to jurisdictional claims in published maps and institutional affiliations.
本文献由“学霸图书馆-文献云下载”收集自网络，仅供学习交流使用。
学霸图书馆（www.xuebalib.com）是一个“整合众多图书馆数据库资源，
提供一站式文献检索和下载服务”的24 小时在线不限IP
图书馆。
图书馆致力于便利、促进学习与科研，提供最强文献下载服务。
图书馆导航：
图书馆首页 文献云下载 图书馆入口 外文数据库大全 疑难文献辅助工具
Related Documents
