Normal formation of a 4-chambered heart is critically

Transcription

1 Right Ventricular Myocardium Derives From the Anterior Heart Field Stéphane Zaffran, Robert G. Kelly, Sigolène M. Meilhac, Margaret E. Buckingham, Nigel A. Brown Abstract The mammalian heart develops from a primary heart tube, which is formed by fusion of bilateral cardiac territories in which myocardial and endothelial cells have already begun to differentiate from splanchnic mesoderm. A population of myocardial precursors has been identified in pharyngeal mesoderm, anterior to the early heart tube. Cell labeling studies have indicated that this novel territory, called the anterior heart field (AHF), gives rise to the myocardial wall of the outflow tract. We now report that not only the myocardium of the outflow tract but also myocardial cells of the embryonic right ventricle are derived from this source. Explants of pharyngeal mesoderm or of the early heart tube were cultured from transgenic mice in which transgene expression marks different regions of the heart. Pharyngeal mesoderm from 5 to 7 somite embryos gives rise to cardiomyocytes with right ventricular and outflow tract identities, whereas the heart tube as this stage has an essentially left ventricular identity. DiI labeling confirms that the early heart tube is destined to contribute to the embryonic left ventricle and indicates that right ventricular myocardium is added from extracardiac mesoderm. Retrospective clonal analysis of the heart at embryonic day (E) 10.5 reveals the existence of a clonal boundary in the interventricular region, which appears during ventricular septation, underlining different origins of the two ventricular compartments. This study demonstrates the differences in the embryological origin of right and left ventricular myocardium, which has important implications for congenital heart disease. (Circ Res. 2004;95: ) Key Words: cardiac development anterior heart field outflow tract right ventricle explants Normal formation of a 4-chambered heart is critically dependent on the early steps of cardiogenesis. These involve the specification and morphogenesis of functional embryonic cardiac chambers, which become remodeled during development of the definitive heart. The initial event in cardiac morphogenesis is the formation of a tubular heart from cardiomyocytes of the cardiac crescent, derived from anterior splanchnic mesoderm (see review 1 ). The linear heart tube is initially orientated along the craniocaudal axis with an anterior outflow (or arterial) pole and a posterior inflow (or venous) pole. Repositioning of the outflow and inflow poles of the heart is brought about by rightward looping, with a dorsoanterior movement of the venous pole behind the developing ventricles (see review 2 ). Cardiac looping in the mouse is initiated at embryonic day (E) 8.25 and is complete by E10.5; during this period, the heart tube grows rapidly in length through the addition of cells to the venous and arterial poles. 3,4 At the arterial pole, precursor cells have been shown to originate from a distinct heart field situated in pharyngeal mesoderm, designated the anterior heart field (AHF). 5 7 Identification of this novel source of myocardial cells raises a number of major questions concerning early events in heart morphogenesis. In particular, it is important to establish the extent of its contribution to the myocardium. In the chick, labeling experiments have suggested that the linear heart tube is composed of the primordia of the apical region of the right ventricle (cranially) and of the left ventricle (caudally). 8 We have investigated this question in the developing mouse heart. Multiple lines of evidence suggest that future right ventricular myocardium is also derived from the AHF. Comparison of the early expression patterns of nlacz transgenes, which provide unique markers of the future right or left ventricles of the midgestation heart, and explant experiments using these transgenic lines, suggest that the linear heart tube contributes predominantly to the embryonic left ventricle and that only the most anterior region contains future right ventricular cells. Explants of pharyngeal mesoderm give rise to myocardial cells with right ventricular as well as outflow tract identity. DiI labeling of cells, followed by embryo culture and analysis of their ventricular location, supports the conclusion that the AHF contributes to the right ventricle. Materials and Methods Transgenic and Targeted Mice The transgenic lines Mlc3f-2, Mlc3f-9, and Mlc1v-24 have been described previously. 5,9,10 -cardiac actin nlaacz1.1/ mice have been Original received September 15, 2003; resubmission received December 23, 2003; revised resubmission received June 14, 2004; accepted June 14, From the Department of Developmental Biology (S.Z., R.G.K., S.M.M., M.E.B.), Pasteur Institute, Paris, France; and St George s Hospital Medical School (N.A.B.), London, UK. Present address for R.G.K. is the Department of Genetics and Development, Columbia University, New York, NY. Correspondence to Stéphane Zaffran, PhD, Department of Developmental Biology, CNRS URA 2578, Pasteur Institute, 25 rue du Dr Roux, Paris, 75015, France. zaffrans@pasteur.fr 2004 American Heart Association, Inc. Circulation Research is available at DOI: /01.RES BE 261

2 262 Circulation Research August 6, 2004 TABLE 1. No. of Explants Analyzed for Transgene Expression Explant No. of Somites Mlc1v-24 Mlc3f-9 Mlc3f-2 Pharyngeal mesoderm AHF 4/5 2 ( ) 3 ( ) 2 ( ) AHF 6/7 5 ( ) 3 ( ) 5 ( ) Heart tube Anterior 5 2 ( ) 3 ( ) 2 ( ) Central 7 2 ( ) 3 ( ) 2 ( ) Anterior 7 2 ( ) 4 ( ) 2 ( ) Results shown in Figures 2 and 5 were consistently obtained with the different transgenes in all experiments listed here. No. of explants and the extent of X-gal staining per beating explant (in parenthesis) are indicated. indicates staining of the complete beating region;, no X-gal labeling. described by Meilhac et al. 11 Mice were maintained on an inverted light-dark cycle to facilitate collection of embryos at the required developmental stages for explant analysis. Transgene expression was monitored by -galactosidase activity using whole-mount and histological revelation as described by Kelly et al. 12 Stained embryos were subsequently sectioned using a cryostat as previously described in Kelly et al. 12 The animals used in this study were mainly produced in the Pasteur Institute Animal Facility, with provision of standard stocks from Janvier (Le Genest St Isle, France). The care and use of laboratory animals followed the guidelines of the French Ministry of Agriculture. Embryonic Explant Cultures The evening of the vaginal plug was taken as E0.5. After isolation of the embryos from the extraembryonic membranes, somites were counted before dissection. The heart tube alone or pharyngeal mesoderm (including pharyngeal endoderm) was dissected using microdissection scissors in PBS and transferred to culture medium in collagen (1%) coated multiwell dishes. Collagen coating promoted attachment of the explants to the culture well. Explants were incubated for 24 to 48 hours at 37 C with 5% CO2 in 250 L of minimal Eagle s medium (MEM) containing 5% horse serum, 100 mmol/l ascorbic acid, and 50 g/ml kanamycin. After incubation, each sample was washed in PBS, followed by fixation in 4% paraformaldehyde in PBS, and three PBS rinses, before staining for -galactosidase activity in X-gal solution as described. 13 Explant experiments were repeated several times for each transgenic line and time point (Table 1). Fluorescent immunohistochemistry was done with the following antibodies: polyclonal anti- -galactosidase HF20 (Molecular Probes; diluted 1:4000) and monoclonal anti-myosin heavy chain (Sigma; 1:100). DiI Injection Embryo culture and DiI labeling were performed as described; 14 labeling at about E8.5 was performed lateroventrally, through the yolk sac, amnion, and pericardial wall. At E8 to E8.25 (4 to 7 somites), DiI was injected into the ventral myocardium of the cardiac crescent or of the heart tube, respectively. Production and Description of nlacz/nlaacz Chimeric Hearts Random clones of myocardial cells were generated spontaneously using the mouse -cardiac actin /nlaacz1.1 line, as described by Meilhac et al. 11 Briefly, a duplication of a portion of the coding region was introduced into the nlacz gene, which interrupts the open reading frame. This construct was targeted to the -cardiac actin locus and is expressed throughout the myocardium. Homologous recombination, which occurs at low frequency, restores the open reading frame of the nlacz gene, resulting in chimeric hearts containing clonally related -galactosidase positive cells that were revealed by X-gal staining as described. Results Early Transgene Markers of Right and Left Ventricular Myocardium The Mlc1v-nlacZ-24 (Mlc1v-24) and Mlc3f-nlacZ-2 (Mlc3f-2) transgenes are expressed in complementary patterns in the midgestation mouse heart (Figure 1a and 1c). The Mlc1v-24 transgene is expressed in the embryonic right ventricle and outflow tract (Figure 1a) as a result of integration of this transgene upstream of the gene encoding fibroblast growth factor 10 (Fgf10), where it is probably the target of an enhancer for regulatory sequences normally acting on the Fgf10 promoter. 5 In contrast, Mlc3f regulatory sequences drive transcription of the Mlc3f-2 transgene (Figure 1c) in the embryonic right atrium and left ventricle, 5,9 whereas the Mlc3f-nlacZ-9 (Mlc3f-9) transgene (Figure 1b) is expressed in both embryonic ventricles but not in outflow tract myocardium. 9 The Mlc1v-24 transgene is expressed in pharyngeal mesoderm mainly situated dorsal to the heart tube at E8.75 (Figure 1g). Sagittal sections through an E8.75 heart, after looping has initiated, reveal transgene expression in arterial pole myocardium, which is immediately adjacent to a layer of pharyngeal mesoderm, corresponding to the dorsal mesocardium, where the transgene is also strongly expressed (Figure 1d, arrowhead). This layer of cells does not express the Mlc3f-9 or Mlc3f-2 transgenes (Figure 1e and 1f; arrowhead) and is in continuity with both the arterial and venous poles of the heart tube (Figure 1d, arrowhead). The Mlc3f-2 transgene is expressed throughout the heart tube with the exception of arterial pole myocardium (Figure 1i, arrowhead), in a complementary profile to the Mlc1v-24 transgene (Figure 1g). The Mlc3f-9 transgene is expressed throughout the linear heart tube at E8.75 (Figure 1h). Pharyngeal Mesoderm Gives Rise to Myocardial Cells With Outflow Tract and Right Ventricular Identities Cell-labeling experiments in chick and mouse embryos have shown that myocardial precursor cells derived from pharyngeal mesoderm participate in the formation of the myocardial wall of the outflow tract. 5 7 In order to examine the myogenic potential of cells in the pharyngeal mesoderm, lying anteriorly and also dorsally to the early heart tube, we cultured this region from embryos of the different transgenic lines described earlier (Figure 1). The Mlc1v-24 transgenic line marks pharyngeal mesoderm as well as, at later stages, the future right ventricular and outflow tract myocardium. In contrast, the other two lines only mark cells in the heart; Mlc3f-9 expression indicates right and left ventricular identity, whereas that of Mlc3f-2 marks the left ventricle. Explants of pharyngeal mesoderm isolated from 6/7 somite stage embryos (E8.25), as shown in Figure 2a through 2c, contained beating foci of myocardial cells after 36 hours of culture (Figure 2d through 2f). To determine the identity of these myocardial cells, we assayed -galactosidase activity in situ. Explants isolated from embryos carrying the Mlc1v-24 transgene were strongly positive. This transgene is already transcribed in pharyngeal mesoderm; however,

3 Zaffran et al Origin of Right Ventricular Myocardium 263 -galactosidase is also detectable in myocardial cells in the explant (Figure 2g). Cardiomyocytes in explants from the Mlc3f-2 transgene were -galactosidase negative, whereas some cells were positive in the Mlc3f-9 explants (Figure 2h and 2i, Table 1). Because Mlc3f-9 and Mlc1v -24 transgenes are coexpressed in the future embryonic right ventricle, but differ in their expression in the outflow tract myocardium of the embryo (Figure 1), these observations suggest that the majority of myocardial cells observed in such explants have an outflow tract identity. Detection of a small number of -galactosidase positive cells in explants derived from Mlc3f-9 embryos (Figure 2h, arrowhead) suggests that a fraction of cardiomyocytes derived from pharyngeal mesoderm at the 6/7 somite stage (E8.5) have a right ventricular identity. To examine this further, explants were excised at the 5 somite stage (E8.25), 4 hours earlier. When these earlier explants were stained after 36 hours of culture, equivalent -galactosidase activity was observed in explants isolated from embryos carrying either the Mlc1v-24 or Mlc3f-9 transgenes (Figure 2j and 2k). No -galactosidase activity was observed in explants isolated from Mlc3f-2 embryos at the 5 somite stage (Figure 2l). Thus, explants cultured from pharyngeal mesoderm isolated at this stage develop a right ventricular identity as indicated by robust expression of the Mlc3f-9 transgene. From these results, we conclude that Figure 1. Comparison of -galactosidase expression profiles in the developing heart for the transgenic mouse lines Mlc1v-24, Mlc3f-9 and Mlc3f-2. a through c, E9.5 hearts in ventral views. a, In Mlc1v-24 transgenic embryos, -galactosidase activity is detected in the outflow tract (OFT) and embryonic right ventricle (RV). Note the sharp boundary between the two ventricular territories. b, In Mlc3f-9 hearts, X-gal staining distinguishes the outflow tract and embryonic right ventricle. -Galactosidase activity is also detected in the left ventricle (LV). c, In Mlc3f-2 embryonic hearts, -galactosidase activity is detected in the left ventricle, whereas the right ventricle and outflow tract are negative. d through f, Sagittal sections showing the expression of nlacz at E8.75. d, -Galactosidase is expressed in the myocardium, at the arterial pole (AP) of the heart tube (ht), and in the pharyngeal mesoderm of the dorsal mesocardium (arrowhead). e and f, -Galactosidase activity is detected in the myocardial wall of the tubular heart with Mlc3f-9 and Mlc3f-2 transgenes, including the venous pole (VP), but excluding the arterial pole (AP) and the dorsal mesocardium (arrowheads). g through i, Ventral views at E8.75; X-gal staining of Mlc3f-9 (h) and Mlc3f-2 (i) transgenic embryos shows -galactosidase activity throughout the tubular heart, with the exception of the arterial pole for Mlc3f-2 (i; arrowhead), whereas for Mlc1v-24 (g; frontal and lateral views) -galactosidase activity is detected at the arterial pole of the heart and in the pharyngeal mesoderm of the dorsal pericardial wall (sm) (heart tube is delimited by a dotted line). i, Straight dotted line indicates approximate plane of the sagittal sections (f, and similar for d and e). pharyngeal mesoderm lying dorsal to the differentiated cells of the early heart tube can give rise to myocardial cells that have an embryonic right ventricular identity and that, subsequently, pharyngeal mesoderm is a source of cells with an outflow tract identity. DiI-labeling experiments, where the dye was injected into pharyngeal mesoderm lying dorsal to the heart tube at E8.5, show that after embryo culture, cells in the outer curvature of the right ventricle are now labeled (Figure 3). This finding is consistent with the results of explant experiments. Cell Lineage Analysis Reveals an Interventricular Boundary Based on the observations shown in Figures 2 and 3, we conclude that right ventricular cardiomyocytes arise from the anterior heart field. As a result, it might be expected that the interventricular region represents a lineage boundary within the developing heart. Retrospective clonal analysis was performed using nlacz/nlaacz chimaeras in which the nlaacz reporter sequence is targeted to the -cardiac actin gene. nlaacz is inactive unless it undergoes a rare intragenic recombination event to yield a clone of nlacz cells, which will be scored as -galactosidase positive when the -cardiac actin gene is expressed (Figure 4). This genetic approach has already contributed to the study of cell behavior in the developing heart. 11,15 Two phases of growth have been

4 264 Circulation Research August 6, 2004 Figure 2. Myocardial differentiation in cultured explants of pharyngeal mesoderm. a through c, Transverse sections of transgenic embryos at the 6/7 somite stage (E8.5), showing the region (delimitated by the lines) from which the explants were isolated. d to l, Explants cultured for 36 hours. d through i, Explants were isolated from the pharyngeal mesoderm of transgenic mouse lines at the 6/7 somite stage (E8.5). At this stage, the region lying under the heart tube was cultured for 48 hours, before detection of -galactosidase activity. d, Phase contrast view of a DAPI-stained explant from the Mlc1v-24 transgenic line. e and f, Cardiomyocytes were observed in contractile foci (within dotted lines). g through i, X-gal staining shows that these cardiomyocytes have a predominantly outflow tract identity because -galactosidase activity is detected mainly in explants isolated from Mlc1v-24 embryos, where coimmunohistochemistry with antibodies to -galactosidase ( -Gal) and myosin heavy chain (MF20) confirms the presence of -galactosidase positive cardiomyocytes (g). Note that a small number of -galactosidase positive cells (arrowhead) are also observed in cultures from Mlc3f-9 embryos (h). j through l, Explants isolated from pharyngeal mesoderm of 4/5 somite stage embryos (E8.25). X-gal staining shows that these cardiomyocytes have a right ventricular identity because -galactosidase activity is detected in both Mlc1v-24 (j) and Mlc3f-9 (k), but not Mlc3f-2 (l) explant cultures. identified: myocardial precursor cells undergo dispersive growth until the time of heart tube formation when myocardial cell growth becomes coherent. 11 We used the nlaacz method to examine the distribution of -galactosidase positive cells in the interventricular region (Table 2). At E10.5, the embryonic right and left ventricles can be distinguished morphologically by the interventricular sulcus, which marks the site of the forming interventricular septum (Figure 4, Figure 3. DiI-labeling of pharyngeal mesoderm. a, Diagram shows the site of DiI injection (asterisk). b, Image of an embryo (lateral view) at the 6/7 somite stage (E8.5) taken immediately after injection (DiI spot). c, Lateral view of the same heart (combined bright field and fluorescence images) after 40 hours of culture, showing Di-labeled cells (red fluorescent) in the outer curvature of the right ventricle. ht indicates heart; BA, branchial arch; LV, left ventricle; RV, right ventricle.

5 Zaffran et al Origin of Right Ventricular Myocardium 265 arrowheads). Four examples of large clusters in the right or left ventricle illustrate that clusters are typically elongated circumferentially in relation to the heart tube, parallel to the interventricular groove (Figure 4a through 4f). The labeled cells do not extend across the two ventricular territories, suggesting that growth of clusters is restricted with respect to the interventricular septum, as confirmed by sections (Figure TABLE 2. No. of -Galactosidase Positive Clones, Examined at E10.5, Which Abut the Interventricular Region Right Ventricle Left Ventricle Large clusters ( 40 cells) Inner curvature 2 1 Outer curvature 3 4 Small clusters (4 40 cells) In this analysis, where 1155 hearts were examined of which 734 had -galactosidase positive cells, large clusters are the most informative because they extend into the ventricle on one side of the septum but do not traverse it. Small clones do not have the same extension, but also show oriented growth with respect to the sulcus. Figure 4. Clusters of clonally related -galactosidase positive cells in the interventricular region. Schematic illustration of the -cardiac actin locus targeted with an nlaacz sequence, which carries a TGA stop codon before a duplication of the nlacz coding region, before (left) and after (right) homologous recombination (red boxes represent -cardiac actin exons, black triangle indicates a residual loxp site after excision of the neomycin cassette; see Meilhac et al 11 ). a through f, Examples of complementary X-gal stained clusters at E10.5, in either the right (a) or left (b) ventricle (ventral views). A boundary delimiting the clusters of -galactosidase positive cells is observed, which corresponds to the physical limit between the ventricles (arrowhead), as observed in sections of these hearts that show -galactosidase positive cells in the outer curvature of either the right (c) or the left (d) side of the interventricular septum. e and f, Examples of clusters of -galactosidase positive cells on the inner curvature in either right (e) or left (f) ventricles adjacent to the interventricular septum (white arrowhead). In neither case does the cluster expand across the septum into the opposite ventricle. 4c and 4d). This observation is consistent with the establishment of a clonal boundary between the two ventricles during the coherent phase of cardiomyocyte growth. Linear Heart Tube Has a Mainly Left Ventricular Identity Our observation that right ventricular cardiomyocytes are added to the linear heart tube from the AHF prompted us to investigate the identity of cardiomyocytes within the heart tube using the same transgenic lines. Explants were isolated from different domains of the heart tube. All cells in explants from the central region of the linear heart tube of Mlc3f-9 and Mlc3f-2 transgenic embryos (E8.5) are -galactosidase positive (Figure 5b and 5c), whereas cells in explants isolated from the same region of the heart tube of Mlc1v-24 embryos were -galactosidase negative (Figure 5a). This result suggests that the central part of the linear heart tube displays a future left ventricular identity. In contrast, explants isolated from the arterial extremity of the linear heart tube, which at this stage lies at right angles to the rest of the tube, express the

6 266 Circulation Research August 6, 2004 Figure 5. Identity of the linear heart tube. Diagrams on the left side illustrate the stage of the embryos used and the part of the heart tube taken for culture (black). The most anterior region (delimited by a dotted line) is shown in the same plane as the rest of the tube, when in fact it is at right angles to it. a through c, Explants were excised from the linear portion of the tubular heart at E8.5 (7 somites). -Galactosidase activity was assayed after 24 to 48 hours of culture. X-gal staining reveals that the linear portion of the embryonic heart tube has a left ventricular identity because -galactosidase activity is observed in both Mlc3f-2 and Mlc3f-9 transgenic explants, whereas no activity is detected in explants from Mlc1v-24 embryos. d through f, Explants dissected from the most anterior portion of the tubular heart at the same stage. -Galactosidase activity is detected in explants derived from Mlc1v-24 and Mlc3f-9 embryos, but not in explants from Mlc3f-2 embryos, indicating a right ventricular identity. g through i, Explants dissected from the most anterior part of this region at the 5 somite stage. X-gal staining reveals a small proportion of cells with right ventricular identity because fewer -galactosidase positive cells are detected in explants derived from Mlc3f-2 embryos (i), compared with those with other transgenes (g and h). Note the smaller number of -galactosidase positive cells in Mlc1v-24 explants (g, arrowhead) compared with explants from Mlc3f-9 embryos (h), in which the transgene is expressed in both future right and left ventricles. Mlc1v-24 and Mlc3f-9 transgenes but not the Mlc3f-2 transgene and therefore display a future right ventricular identity (Figure 5d through 5f). To determine whether future right ventricular cardiomyocytes are already present in the forming tube at an earlier stage, we performed explant experiments with parts of the heart tube from 5 somite stage embryos (E8.25). Explants isolated from the anterior cardiac region of Mlc1v-24 transgenic embryos show that this is the case (Figure 5g), although some future left ventricular cells are also present (Figure 5i), as also demonstrated by the complete X-gal staining of explants from the Mlc3f-9 transgenic line (Figure 5h). To confirm these observations, based on transgene expression profiles, we performed a series of DiI-labeling experiments. The heart was labeled by DiI injection, followed by embryo culture for up to 40 hours, during which time cardiac looping took place (Figure 6). When the medial portion of the ventral myocardium of the cardiac crescent was injected with DiI at the E8.25 stage (Figure 6a and 6b), labeled cells were found in the left ventricle adjacent to the interventricular septum (Figure 6c). Interestingly, a similar result was obtained when the anterior region of the heart tube was labeled at the E8.5 stage (Figure 6d through 6f). When injections were made into the right side of the midportion of the heart tube, after looping has initiated at the E8.5 stage (Figure 6g and 6h), DiI-labeled cells were found in the outer curvature of the interventricular region (Figure 6i). This result shows that the major part of the linear heart tube is fated to give rise to the left ventricle and supports the conclusion, drawn from the explant experiments, that the main identity of the linear heart tube in the mouse is that of future left ventricular myocardium. Discussion We have investigated the extent to which the AHF contributes to the embryonic heart. Using explant culture coupled with regionalized transgene markers and DiI-injection, we have documented the progressive acquisition of myocardial cells with right ventricular and outflow tract identity as the heart tube, initially composed almost entirely of prospective left ventricular myocardium, elongates anteriorly. A major part of the embryonic heart, including the right ventricle as well as the outflow tract, is therefore derived from precursor cells situated in pharyngeal mesoderm, initially lying dorsally and then anteriorly to the cardiac tube. This finding was suggested by the expression profile of the Mlc1v-24 transgenic mouse line, where insertion of the transgene adjacent to Fgf10, leads to transcription of the nlacz reporter in Fgf10 expressing cells of the AHF, and detection of

7 Zaffran et al Origin of Right Ventricular Myocardium 267 Figure 6. DiI-labeling of the developing heart at different stages. a, d, and g, Diagrams illustrate embryos showing the site of DiI injection (*). Images of embryos (ventral view) at the 4 somite (b), 6 somite (e), and 8 somite (h) stages taken immediately after injections (DiI spot). c and f, Ventral views of the embryonic hearts shown in a and b and d and e after embryo culture, showing DiI-labeled cells in the left ventricle adjacent to the interventricular sulcus (arrowheads). i, Embryonic hearts seen in g and h after 40 hours of culture, showing labeled cells in the outer curvature in the region of the interventricular sulcus (arrowhead). cc indicates cardiac crescent; ht, heart tube; LV, left ventricle; RV, right ventricle. -galactosidase protein, but not transcript, in the myocardium of the right ventricle, as well as the outflow tract. 5 A recent publication on the Islet 1 mutant mouse shows expression of this gene in pharyngeal mesoderm, lying both anteriorly and dorsally to the developing heart tube, in a profile that overlaps with that of the Mlc1v-24 transgene. 16 A lineage tracing study with an Islet 1 Cre and the Rosa 26 lacz mouse suggested that these cells contribute to both outflow tract and right ventricular myocardium. Interestingly, there was also a contribution to the venous pole. 16 Our observation of nlacz/nlaacz chimeric hearts at E10.5 reveals that labeled clusters of clonally related cells are restricted to either left or right ventricular compartments. Furthermore, differential patterns of oriented clonal cell growth show that right and left ventricular cells display differential behavior probably in response to distinct developmental programs during heart formation. 15 We have recently completed an extensive retrospective clonal analysis of myocardial cells at E8.5, which demonstrates the existence of two cell lineages: left ventricular myocardium is derived exclusively from the first lineage, whereas the outflow tract myocardium, together with part of the right ventricular (and atrial) myocardium is derived from the second lineage. 17 This lineage analysis is in keeping with the results reported in this study on the predominantly left ventricular identity of the early heart tube. However, the observation that some clones span future right and left ventricular territories within the E8.5 heart tube suggests that these are not completely restricted during the earliest stages of heart development, and that the right/left ventricular boundary is established later. Some very large clones extend over the entire length of the cardiac tube pointing to an early common progenitor of the two lineages. The differences between progenitor cell populations that contribute to the heart thus may be more critically temporal, rather than spatial. This clonal analysis, together with the observation on Islet 1 and the initial location of cardiac cell progenitors in pharyngeal mesoderm lying dorsal to the cardiac tube, will necessitate reconsideration of the AHF nomenclature. Using in vivo labeling studies in the chick, de la Cruz and colleagues 3,8 mapped the fate of the myocardial component of the linear heart tube. These studies demonstrated that it is composed of two regions, the primordium of the right ventricle, situated anteriorly and the primordium of the left ventricle situated posteriorly. This is in contrast to our results that show that the early heart tube in the mouse embryo is predominantly fated to give rise to the embryonic left ventricle, with the exception of a few cells in the most anterior region. This apparent discrepancy between birds and mammals may be due to differences in the timing of heart tube formation with respect to that of the intraembryonic coelom. In addition, looping is initiated later in the chick embryo than in the mouse and therefore the linear heart tube stage is more evident in the former. The contribution of pharyngeal mesoderm from the secondary, 7 or the anterior heart field, 6 to the avian cardiac tube is limited to outflow tract myocardium. At the molecular level, the transcriptional regulation of the Mlc3f-2 transgene, and other transgenes such as Mlc2v-lacZ, 18 underlines the difference between left and right embryonic ventricles in the mouse embryo. In contrast, a gene such as dhand, which only becomes restricted to the right ventricle later,

8 268 Circulation Research August 6, 2004 does not appear to reflect the different cellular origins of these compartments. 19 The Tbx5 gene, encoding a T-box transcription factor, is interesting in this context. It is initially expressed throughout the cardiac crescent and in the linear tube, with subsequent expression predominantly in the left ventricle (and atria). 20 Tbx5 deficiency in homozygous mice causes severe hypoplasia of posterior segments in the developing heart, with an apparently normal right ventricle and outflow tract. 21 Furthermore, in chick or mouse embryos an ectopic expression of Tbx5 in right ventricular precursor cells gives rise to a heart with a single ventricular chamber with left identity. 22 These observations suggest that Tbx5 is a gene required in the linear heart tube, which has an essentially left ventricular identity, but not in cardiomyocytes derived from the AHF, which contribute to the right ventricle. Expression of Tbx5 in the atria and single ventricular chamber of the amphibian heart is essential for cardiogenesis. 23 In species with a right ventricular chamber and separate pulmonary circulatory system, Tbx5 function within the left ventricle (and atria) appears to be maintained. 21 The conclusion that the right ventricular, as well as outflow tract, myocardium is derived from pharyngeal mesoderm has implications for the understanding of congenital heart disease. The spectrum of congenital heart defects found in human patients illustrates the modular development of the heart, because a particular region, rather than the heart as a whole, is frequently affected. 24 Ventricular hypoplasia of either ventricle is one of the more severe defects encountered in pediatric cardiology. 25 Hypoplasia resulting in abnormal physiology of a single ventricle is typical and the unaffected ventricle in these cases is usually morphologically and physiologically normal. This observation underscores the differences in developmental program between right and left ventricular chambers and supports the hypothesis that right and left ventricular morphogenesis is regulated by different genetic pathways reflecting differences in their embryological origins. Defects in ventricular septation are a second class of severe congenital heart defects. Such defects can now be considered to arise at the interface between cardiomyocytes derived from the cardiac crescent and those derived from the pharyngeal mesoderm of the anterior heart field. Acknowledgments The laboratory of M.B. is supported by the Pasteur Institute and the CNRS and a grant from the ACI Integrative Biology Program of the French Research Ministry. S.M. was supported by a fellowship from the French Research Ministry and the University of Paris. R.K. is an INSERM research fellow. S.Z. is a CNRS research fellow and acknowledges prior support from the FRM. The laboratory of N.B. is supported by the British Heart Foundation Program Grant RG/03/ 012. We are grateful to Dr Diego Franco for helpful discussions and Dr Andrew Munk for advice on explant culture. We thank Emmanuel Pecnard for technical assistance. References 1. Yutzey KE, Kirby ML. Wherefore heart thou? Embryonic origins of cardiogenic mesoderm. Dev Dyn. 2002;223: Harvey RP. Organogenesis: patterning the vertebrate heart. Nat Rev Genet. 2002;3: de la Cruz MV, Sanchez Gomez C, Arteaga MM, Arguello C. Experimental study of the development of the truncus and the conus in the chick embryo. J Anat. 1977;123: Viragh S, Challice CE. Origin and differentiation of cardiac muscle cells in the mouse. J Ultrastructure Res. 1973;42: Kelly RG, Brown NA, Buckingham ME. The arterial pole of the mouse heart forms from Fgf10-expressing cells in pharyngeal mesoderm. Dev Cell. 2001; 1: Mjaatvedt CH, Nakaoka T, Moreno-Rodriguez R, Norris RA, Kern MJ, Eisenberg CA, Turner D, Markwald RR. The outflow tract of the heart is recruited from a novel heart-forming field. Dev Biol. 2001;238: Waldo KL, Kumiski DH, Wallis KT, Stadt HA, Hutson MR, Platt DH, Kirby ML. Conotruncal myocardium arises from a secondary heart field. Development. 2001;128: de la Cruz MV, Sanchez-Gomez C, Palomino MA. The primitive cardiac regions in the straight tube heart (stage 9) and their anatomical expression in the mature heart: an experimental study in the chick embryo. J Anat. 1989; 165: Franco D, Kelly R, Lamers W, Buckingham M, Moorman A. Regionalized transcriptional domains of myosin light chain 3f transgenes in the embryonic mouse heart: morphogenetic implications. Dev Biol. 1997;188: Kelly R, Zammit P, Mouly V, Butler-Browne G, Buckingham M. Dynamic left/right regionalisation of endogenous myosin light chain 3F transcripts in the developing mouse heart. J Mol Cell Cardiol. 1998;30: Meilhac SM, Kelly RG, Rocancourt D, Eloy-Trinquet S, Nicolas JF, Buckingham ME. A retrospective clonal analysis of the myocardium reveals two phases of clonal growth in the developing mouse heart. Development. 2003;130: Kelly R, Alonso S, Tajbakhsh S, Cossu G, Buckingham M. Myosin light chain 3F regulatory sequences confer regionalized cardiac and skeletal muscle expression in transgenic mice. J Cell Biol. 1995;129: Sanes JR, Rubenstein JLR, Nicolas JF. Use of a recombinant retrovirus to study post-implanation cell lineage in mouse embryos. EMBO J. 1986;5: Franco D, Kelly R, Moorman AF, Lamers WH, Buckingham M, Brown NA. MLC3F transgene expression in iv mutant mice reveals the importance of left-right signalling pathways for the acquisition of left and right atrial but not ventricular compartment identity. Dev Dyn. 2001;221: Meilhac SM, Esner M, Kerszberg M, Moss JE, Buckingham ME. Oriented clonal cell growth in the developing mouse myocardium underlies cardiac morphogenesis. J Cell Biol. 2004;164: Cai CL, Liang X, Shi Y, Chu PH, Pfaff SL, Chen J, Evans S. Isl1 identifies a cardiac progenitor population that proliferates prior to differentiation and contributes a majority of cells to the heart. Dev Cell. 2003;5: Meilhac S, Esner M, Kelly R, Nicolas JF, Buckingham M. The clonal origin of myocardial cells in different regions of the embryonic mouse heart. Dev Cell. May;6: Ross RS, Navankasattusas S, Harvey RP, Chien KR. An HF-1a/HF- 1b/MEF-2 combinatorial element confers cardiac ventricular specificity and established an anterior-posterior gradient of expression. Development. 1996; 122: Srivastava D, Thomas T, Lin Q, Kirby ML, Brown D, Olson EN. Regulation of cardiac mesodermal and neural crest development by the bhlh transcription factor, dhand. Nat Genet. 1997;16: Bruneau BG, Logan M, Davis N, Levi T, Tabin CJ, Seidman JG, Seidman CE. Chamber-specific cardiac expression of Tbx5 and heart defects in Holt-Oram syndrome. Dev Biol. 1999;211: Bruneau BG, Nemer G, Schmitt JP, Charron F, Robitaille L, Caron S, Conner DA, Gessler M, Nemer M, Seidman CE, Seidman JG. A murine model of Holt-Oram syndrome defines roles of the T-box transcription factor Tbx5 in cardiogenesis and disease. Cell. 2001;106: Takeuchi JK, Ohgi M, Koshiba-Takeuchi K, Shiratori H, Sakaki I, Ogura K, Saijoh Y, Ogura T. Tbx5 specifies the left/right ventricles and ventricular septum position during cardiogenesis. Development. 2003;130: Horb ME, Thomsen GH. Tbx5 is essential for heart development. Development. 1999;126: Fishman MC, Olson EN. Parsing the heart: genetic modules for organ assembly. Cell. 1997;91: Srivastava D, Olson EN. A genetic blueprint for cardiac development. Nature. 2000;407: