Spatiotemporally separated cardiac neural crest subpopulations that target the outflow tract septum and pharyngeal arch arteries.код для вставкиСкачать
THE ANATOMICAL RECORD PART A 275A:1009 –1018 (2003) Spatiotemporally Separated Cardiac Neural Crest Subpopulations That Target the Outﬂow Tract Septum and Pharyngeal Arch Arteries MARIT J. BOOT, ADRIANA C. GITTENBERGER-DE GROOT, LIESBETH VAN IPEREN, BEEREND P. HIERCK, AND ROBERT E. POELMANN* Department of Anatomy and Embryology, Leiden University Medical Center, Leiden, The Netherlands ABSTRACT We used lacZ-retrovirus labeling combined with neural crest ablation in chick embryos to determine whether the cardiac neural crest cells constitute one group of multipotent cells, or they emigrate from the neural tube in time-dependent groups with different fates in the developing cardiovascular system. We demonstrated that early-migrating cardiac neural crest cells (HH9 –10) massively target the aorticopulmonary septum and pharyngeal arch arteries, while the late-migrating cardiac neural crest cells (HH12) are restricted to the proximal part of the pharyngeal arch arteries. These results suggest a prominent role for early-migrating cells in outﬂow tract septation, and a function for late-migrating cells in pharyngeal arch artery remodeling. We demonstrated in cultures of neural tube explants an intrinsic difference between the early and late populations. However, by performing heterochronic transplantations we showed that the late-migrating cardiac neural crest cells were not developmentally restricted, and could contribute to the condensed mesenchyme of the aorticopulmonary septum when transplanted to a younger environment. Our ﬁndings on the exact timing and migratory behavior of cardiac neural crest cells will help narrow the range of factors and genes that are involved in neural crest-related congenital heart diseases. Anat Rec Part A 275A:1009 –1018, 2003. © 2003 Wiley-Liss, Inc. Key words: chick embryo; heart development; migration; ablation; neural tube cultures; heterochronic transplantation The neural crest in the early avian embryo is formed at the dorsal midline as the neural folds fuse. The migratory behavior of neural crest cells is characteristic of each axial level, and is dependent on the developmental stage of the embryo (Tosney, 1982). The region of the neural crest between the midotic placode and the caudal limit of somite 3 has been called the cardiac neural crest because of its involvement in cardiac development (Kirby et al., 1985). In studies of quail-chick chimeras, cardiac neural crest cells were observed in the aorticopulmonary septum, branchial arch arteries, truncus arteriosus, and cardiac ganglia (Le Lièvre and Le Douarin, 1975; Kirby et al., 1983). In studies using retroviral labeling, neural crest cells were shown to be among the ﬁrst cells to differentiate into primary smooth muscle cells of the arch arteries, and to contribute later in development to the adventitia and media of the pharyngeal arch arteries (Bergwerff et al., 1998). Preferential labeling has been detected along the medial side of the vessel walls of the aorta and pulmonary © 2003 WILEY-LISS, INC. trunk, continuing into the condensed mesenchyme and prongs of the aorticopulmonary septum (Bergwerff et al., 1998; Waldo et al., 1998). The condensed neural crest cells of the aorticopulmonary septal complex and the prongs disperse into scattered neural crest cells within the conal cushions, extending to the free rim of the right ventricular outﬂow tract septum (Poelmann et al., 1998; Waldo et al., 1998). The neural crest cells of the prongs display abun- *Correspondence to: Prof. Dr. R.E. Poelmann, Dept. of Anatomy and Embryology, Leiden University Medical Center, P.O. Box 9602, 2300 RC Leiden, The Netherlands. Fax: ⫹31.71.5276680. E-mail: R.E.Poelmann@lumc.nl Received 29 April 2003; Accepted 15 August 2003 DOI 10.1002/ar.a.10099 1010 BOOT ET AL. dant apoptosis, while only a low number of neural crest cells located in the pharyngeal arch artery vessel walls are apoptotic (Poelmann et al., 1998, 2000). In other studies using retroviral labeling, the contribution of the cardiac neural crest to the autonomous nervous system (Verberne et al., 1998, 2000), as well as to regions of the prospective conduction system and a mesenchymal population at the venous pole of the heart (Poelmann and Gittenberger-de Groot, 1999), were described in detail. The importance of the cardiac neural crest population for normal heart development was conﬁrmed in studies of chick embryos in which this population was surgically removed. These embryos displayed cardiovascular malformations, such as persistent truncus arteriosus, ventricular septal defects, inﬂow anomalies, and interruption or persistence of the aortic arch arteries (Bockman et al., 1989; Kirby and Waldo, 1990, 1995). The cardiovascular phenotypic features described in embryos in which the neural crest was ablated are similar to those found in certain human diseases, such as DiGeorge syndrome and velocardiofacial syndrome (Schiafﬁno et al., 1999). This has led to the proposal that DiGeorge syndrome is a neural crest-related disease, the phenotype of which results from the absence of a gene copy or from diminished doses of speciﬁc gene products at critical time points during early development (Goldmuntz and Emanuel, 1997; Garg et al., 2001; Kochilas et al., 2002). It is necessary to understand the exact differentiation (Thomas et al., 1998), timing, and migratory behavior of cardiac neural crest cells in order to elucidate the underlying genetic mechanisms and environmental messages. In a previous study (Boot et al., 2003), we demonstrated that the exact timing of cardiac neural crest cell emigration from the neural tube in chick embryos was between Hamburger and Hamilton (1951) stages HH9 and HH13–, and that at HH14 and HH15 no cells were emigrating from the dorsal or ventral side of the neural tube between the level of the otic placode and the caudal limit of somite 3. The cardiac neural crest cells have been regarded as one group of cells with the capacity to contribute to various aspects of the developing cardiovascular system, ranging from the tunica media of the aortic arch arteries to the condensed mesenchyme in the aorticopulmonary septum of the outﬂow tract (Kirby, 1993). However, a direct correlation between the occurrence of a persistent truncus arteriosus and morphological abnormalities of the aortic arch arteries has not been described, even though these abnormalities have been found both together and separately after ablation of the neural crest (Kirby and Waldo, 1990). Whether this is due to variations in the execution of the ablation technique, or reﬂects the fact that we are dealing with different subpopulations of cardiac neural crest cells is the subject of this study. At the cardiac level, no division into early and late neural crest cell migration patterns, nor any studies on limited developmental potential have been described. The ﬁrst focus of this study was the migratory behavior of time-dependent migratory groups within the cardiac neural crest population, which we examined by marking the neural crest cells with a retroviral construct harboring the LacZ reporter gene (Bergwerff et al., 1998; Poelmann et al., 1998). Retrovirus was injected into the neural tube, either early (HH9 –10) or late (HH12) during cardiac neural crest cell migration, and the contribution of the total cardiac neural crest population was compared to the con- tribution of the late-migrating neural crest only. By combined early neural crest cell labeling and removal of the late cardiac neural crest cells, the contribution of the early-migrating cardiac neural crest cell population was studied separately. These embryos also showed the effects of ablation of late-migrating cardiac neural crest cells on cardiovascular development. The second focus of this study was to determine whether there are intrinsic differences between early- and latemigrating cardiac neural crest cells. Therefore, we cultured early (HH10) and late (HH12) neural tube explants, and compared the in vitro migration and differentiation patterns of the emigrating cardiac neural crest cells. The third focus of this study was the inﬂuence of early and late embryonic environments on neural crest cell migration, which we studied by performing heterochronic transplantations. Late (HH12) quail cardiac neural crest cells were transplanted into an early (HH10) chick embryo in which the cardiac neural crest had been surgically removed. Using retroviral labeling, in vitro neural tube explant cultures, and quail-chick chimera techniques, we demonstrated that the cardiac neural crest consists of early- and late-migrating subpopulations that target the outﬂow tract septum and the pharyngeal arch arteries, respectively. MATERIALS AND METHODS Retrovirus The replication-incompetent spleen-necrosis retrovirus that was used in this study contains a bacterial lacZ gene, as described by Mikawa et al. (1991). The retrovirus was produced in the canine cell line D17.2G/CXL and secreted into the culture medium (Iscove’s DMEM containing 5% fetal calf serum and 1% penicillin/streptomycin). After the cell debris was removed, the virus was harvested by centrifugation (17,000 g for 21⁄2 hr) of the supernatant. The pellet was resuspended in a small volume of medium containing 100 g/ml polybrene to improve transduction, and 0.25 g/ml indigo carmine blue for visualization of the virus solution during in ovo injection. The titer of the retrovirus was determined on rat R2 ﬁbroblasts. The titers of the virus solutions used in these experiments varied between approximately 1–5 ⫻ 106 transducing units per milliliter. Infection of Chicken Embryos Fertilized, speciﬁed pathogen-free White Leghorn eggs were incubated at 37°C for 33– 46 hr and windowed at HH9 –12. The retrovirus solution was drawn into a glass micropipette mounted on a micromanipulator, and then carefully injected into the lumen of the neural tube using a programmable microinjector (IM-300 Narishige, Japan). Stage HH9 –12 embryos were injected into the neural tube at somite levels 4 – 6, ﬁlling the neural tube in the anterior direction until the virus solution reached the otic placode level, to label the complete cardiac neural crest cell region. In addition, part of the cranial and trunk neural crest cells were labeled by diffusion of the retrovirus; however, these cells do not contribute to the cardiovascular system (Kirby et al., 1985; Artinger and Bronner-Fraser, 1992; Baker et al., 1997) . After the embryos were injected, they were washed with Locke’s solution. The eggs were sealed with Scotch tape and returned to the incubator at 37°C for further development. CARDIAC NEURAL CREST SUBPOPULATIONS Labeling of Early Neural Crest and Ablation of Late Cardiac Neural Crest In double experiments, embryos were injected with retrovirus at HH10 as described above, and subsequently reincubated for another 3– 4 hr at 37°C until they reached HH11. At this stage the dorsal side of the neural tube was microsurgically removed between the otic placode and the caudal part of somite 3 (cardiac neural crest), using a sharpened tungsten needle. The eggs were sealed and the embryos survived until HH13–15 or HH30 –31. Embryo Fixation and ␤-Galactosidase Staining The embryos were allowed to survive for 8 hr to 7 days after injection. HH11–28 embryos were immersion ﬁxed in toto. HH29 –32 embryos were decapitated and the thorax was opened. In HH33–34 embryos the thorax was isolated to optimally expose the heart and vessels to 2% paraformaldehyde in 0.2 M phosphate buffer for 1 hr, followed by extensive rinsing in PBS. Subsequently, the embryos were stained for the presence of lacZ by immersion in X-gal solution (PBS containing 5 mM potassium ferricyanide, 5 mM potassium ferrocyanide, 2 mM MgCl2, and X-Gal (5bromo-4-chloro-3 indolyl ␤-D-galactopyranoside) for 2–5 hr at 37°C while being shaken. The stained embryos were macroscopically evaluated for successful blue labeling of lacZ-expressing cells. Infection grades vary among embryos of the same infection stage; therefore, we only selected intensely stained embryos. Similar numbers of intensely stained embryos were found in each group. The group with HH9 or HH10 injected embryos was referred to as the “early-injected group,” with labeling of early- and late-migrating cardiac neural crest cells. HH12 injected embryos were referred to as the “late-injected group,” with labeling of the late-migrating cardiac neural crest cells. The following conditions were studied: HH9 or 10 injection (n ⫽ 19), HH10 injection combined with HH11 ablation (n ⫽ 5), HH11 injection (n ⫽ 8), and HH12 injection (n ⫽ 18). Heterochronic Transplantation Fertilized eggs from White Leghorn chickens and Japanese quails were incubated for 38 and 44 hr, respectively. The eggs were windowed, and HH10 chick embryos and HH12 quail embryos were selected according to the number of somites. Cardiac neural crest cell migration patterns in chick and quail embryos are very similar; therefore, quail-chick chimeras have been used in several studies to describe neural crest cell migration (Kirby et al., 1983; Verberne et al., 2000). For our heterochronic transplantation experiment, the chick cardiac neural crest region was removed with a tungsten needle. The cardiac neural crest was excised from an HH12 quail embryo and transplanted into the graft site of an HH10 chick embryo. The eggs were sealed and the embryos (n ⫽ 5) survived until HH30, followed by overnight immersion ﬁxation in 4% paraformaldehyde in 0.1 M phosphate buffer and parafﬁn embedding. In Vitro Culture System HH10 and HH12 chick embryos were used for early and late neural tube cultures, respectively. The embryos were taken from the yolk sac and washed in Locke’s solution. The neural tube region from the mesencephalon to the 1011 caudal level of the third somite pair was excised. The somites and mesenchymal tissue were removed by placing the neural tube region in a collagenase solution (0.5% in Locke’s solution) for 10 min. Subsequently, the neural tube region was washed with medium and transferred onto a ﬁbronectin-coated (20 g/ml) coverslip in a 24-well plate and grown in 1 mL Medium 199 containing 10% fetal calf serum, 0.5% gentamicin, 1% penicillin/streptomycin, 1% ITS (Invitrogen Corporation, Paisley, UK). After 24 or 48 hr, the cultures were ﬁxed for 30 min in 4% paraformaldehyde/0.1 M phosphate buffer and stored in 70% ethanol at 4°C before the early (n ⫽ 24) and late (n ⫽ 22) neural tube explant cultures were immunohistochemically processed. Immunohistochemistry The experimental embryos were embedded in parafﬁn, and serial sections (5 m) were made. The ﬁxed explant cultures and alternate parafﬁn sections were immunohistochemically stained for muscle actin using the HHF35 antibody (DAKO, Glostrup, Denmark) (Tsukada et al., 1987) and the HNK-1 epitope (Hybridoma Bank, Iowa City, IA) (Abo and Balch, 1981) to mark muscle cells or neural crest and differentiated nerve tissues (Poelmann et al., 1998), respectively. Quail-derived cells were identiﬁed using the QCPN antibody, a quail nuclear marker (Hybridoma Bank). Overnight incubations were performed with the primary antibodies diluted (HHF35 1:1,000, HNK-1 1:10, and QCPN 1:1 for sections; and HHF35 1:500 and HNK-1 1:50 for cultures) in PBS 0.05% Tween-20 and 1% chicken egg albumin. After the sections and cultures were rinsed, they were incubated for 2 hr with peroxidaseconjugated rabbit anti-mouse antibodies (1:250, DAKO). They were then washed, and a DAB-H2O2 procedure was performed with 0.04% diaminobenzidine tetrahydrochloride, 0.006% H2O2 in 0.05 M Tris-maleic acid (pH 7.6) for 10 min. The sections and cultures were counterstained with Mayer’s hematoxylin, dehydrated, and mounted in Entellan. RESULTS Comparison of the Migration Patterns of the Total Cardiac Neural Crest Cell Population and the Late-Migrating Cardiac Neural Crest Cell Population Embryos were injected with a retrovirus containing the lacZ marker gene into the neural tube lumen at HH9 –12. Embryos injected at HH9 or HH10 formed the early-injected group, in which the lacZ gene of the retrovirus was integrated into replicating neuroepithelial cells that gave rise to the total population of cardiac neural crest cells. Embryos injected at HH12 comprised the late-injected group, in which the late-migrating cardiac neural crest cells that leave the neural tube around HH12 were exclusively labeled. The migration patterns of labeled cardiac neural crest cells in the early- and late-injected groups were very similar at the ﬁrst days (HH15/16, HH22, and HH26) following retroviral labeling. Both the early- (Fig. 1a) and lateinjected (Fig. 1e) embryos displayed lacZ-labeled cardiac neural crest cells outside the dorsal side of the neural tube and in the mesenchyme caudal to the otic placode. The lacZ-labeled cells were observed in the circumpharyngeal region around the dorsal aorta as strings of HNK-1 posi- Fig. 1. Injection of lacZ retrovirus into the neural tube lumen followed by ␤-galactosidase staining. a– d: Embryos injected at HH9 or HH10 survived until HH11, HH15, or HH30. a: Blue-labeled neural crest cells (arrows) outside the neural tube at the level of the cardiac neural crest. b: Labeled neural crest cells are located in the III (arrows) and IV pharyngeal arches. Other cells, supposedly of neural crest origin, stained brown with the HNK-1 antibody. c: Labeled neural crest cells are located in the proximal and distal parts (arrows) of the pharyngeal arch arteries. d: High numbers of labeled neural crest cells are found in the condensed mesenchyme (CM) of the aorticopulmonary septum. e– h: Embryos injected at HH12 survived until HH16, HH30, or HH33 to trace the latemigrating cardiac neural crest cells. e: Labeled neural crest cells are located in the head mesenchyme cranial and caudal of the otic placode (OP). f: Late-migrating cardiac neural crest cells just posterior to the otic placode. Note that the blue cells are double-stained for HNK-1 (arrow). g: Many blue cells in the proximal, actin-negative part of the pharyngeal arch arteries III, IV, and VI. h: Blue cells at the peripheral, ventral side of the aortic wall. Note the absence of lacZ-positive cells in the condensed mesenchyme (CM) of the aorticopulmonary septum. i–m: Embryos that were injected at HH10, and ablated at HH11 to trace the early-migrating cardiac neural crest cells, survived until HH31. i: The labeled cells are scattered over the pharyngeal arch arteries, and are concentrated in the condensed mesenchyme and prongs of the aorticopulmonary septum (arrows). j: The distal part of the left third pharyngeal arch artery has an irregularly-shaped lumen and shows labeled early-migrating cardiac neural crest cells. k: A segment of the vessel wall of the right sixth arch artery is abnormally organized (between arrows). l: The labeled earlymigrating neural crest cells are evenly distributed over both the actinnegative and -positive parts of the right sixth arch artery (RVI). m: The centrally located condensed mesenchyme (CM) of the aorticopulmonary septum contains a large number of blue-stained early-migrating cardiac neural crest cells. Bar ⫽ (j) 75 m, and (a–i, k–m) 150 m. Abbreviations: Ao, aorta; CM, condensed mesenchyme; DAo, dorsal aortae; N, notochord; OP, otic placode; P, pulmonary trunk; R, rhombencephalon; S1, somite 1; II, III, IV, VI, second, third, fourth, and sixth pharyngeal arch arteries; L, left; R, right. CARDIAC NEURAL CREST SUBPOPULATIONS tive cells, and in the mesenchyme of the arches (Fig. 1b and f). However, at stage HH30 the patterns of the labeled cardiac neural crest cells in the early- and late-injected groups were strikingly different. In the ﬁrst group of embryos, which were injected at HH9 or 10, the labeled neural crest cells were scattered over the vessel walls of the third, fourth, and sixth pharyngeal arch arteries (Fig. 1c) at HH30. The cells were observed both in the actinnegative proximal part of the pharyngeal arch arteries close to the aortic sac region, and in the more distal, actin-positive part of the pharyngeal arch arteries close to the descending aorta (Fig. 1c). The cardiac neural crest cells did not contribute to the descending aorta. The labeled neural crest cells were incorporated into the periendothelial layer, as well as the peripheral layers of the vessel walls. Furthermore, the labeled cells contributed intensely to the condensed mesenchyme and the prongs of the aorticopulmonary septum (Fig. 1d). In contrast, the second group of embryos injected at HH12 demonstrated that the population of late-migrating cardiac neural crest cells was located almost exclusively in the proximal part of the third, fourth, and sixth pharyngeal arch arteries at HH30 –HH33 (Fig. 1g). The position of the late-migrating cardiac neural crest cells colocalized largely with the actin-negative (HHF35) boundaries at these stages of development. The late-migrating neural crest cells were not limited to a speciﬁc position in the vessel wall, but were encountered in both the periendothelial layer and the more peripheral layers. The late-migrating cardiac neural crest cells contributed to the medial side of the pulmonary trunk and the ascending aorta, as well as to the ventral side of the ascending aorta (Fig. 1h). The latter group of positive cells extended with a low number of cells into the outﬂow tract cushion tissue. Interestingly, the late-migrating cardiac neural crest cells were not found in the prongs, and were nearly absent in the condensed mesenchyme of the aorticopulmonary septum (Fig. 1h). By subtracting the expression pattern of the late-migrating cardiac neural crest cells (injection HH12) from the expression pattern of the total cardiac neural crest cell population (injection HH9 –10), we expected the earlymigrating cardiac neural crest cells to migrate to the pharyngeal arch arteries and, more speciﬁcally, to the condensed mesenchyme of the aorticopulmonary septum. Migration Pattern of the Early-Migrating Cardiac Neural Crest Cell Population We exclusively studied the early-migrating cardiac neural crest cells that leave the neural tube around HH10 by injecting retrovirus into the neural tube lumen at HH10, reincubating the egg, and subsequently ablating the latemigrating cardiac neural crest at HH11. At HH30 –31 the distribution of the late-migrating cardiac neural crest and the effect of the late-ablation on the heart and pharyngeal arch arteries (Fig. 1i) were assessed. The ablation of the late-migrating cardiac neural crest resulted in malformed pharyngeal arch arteries III, IV, and VI with an irregular lumen structure (Fig. 1j) and abnormal vessel walls in which the layers were stacked abnormally (Fig. 1k). Furthermore, the proximal part of the left third arch artery was hypoplastic. It must be stressed that the aorticopulmonary septum was formed normally in these late-ablated embryos (Fig. 1m). The labeled early-migrating neural crest cells were present in the pharyngeal arch mesen- 1013 chyme and scattered in the media of the third, fourth, and sixth pharyngeal arch arteries. Cells were noted in the periendothelial layer and the more peripheral layers of the arch artery vessel walls in both the proximal parts (which are actin-negative) and the more distal parts (which are actin-positive) (Fig. 1l) of the arteries. Early neural crest cells were found at the medial side of the pulmonary trunk and ascending aorta immediately downstream of valve level. Furthermore, they contribute heavily to the condensed mesenchyme of the aorticopulmonary septum (Fig. 1m), with a few cells scattered in the prongs. The experimental procedure and the results are summarized in Fig. 2a and b. It must be noted that the differences in pharyngeal arch artery structure shown in Fig. 2a and b are partly due to the difference in developmental stage (HH30 vs. HH31). However, the irregular shape of the pharyngeal arch arteries and the hypoplasia of the LIII were not related to the developmental stage; rather, they were caused by the ablation of the late-cardiac neural crest. In Vitro Cultures of Early and Late Neural Tube Explants We evaluated the intrinsic differences between the total neural crest cell population and the late cardiac neural crest cell population by culturing early (HH10) and late (HH12) neural tube explants. In the ﬁrst hours after the explants were placed in culture medium, we observed proliferating neuroepithelial cells (identiﬁed by their distinct morphology of compactly organized epithelial cells) and migrating neural crest cells (recognizable by their protrusions and web-shaped appearance). After 24 hr in culture, the early and late explants all showed high numbers of neural crest cells. Staining for HNK-1 showed that in the early-explant cultures almost all neural crest cells were positive for HNK-1 (Fig. 3a), whereas in the lateexplant cultures a number of cells with the typical neural crest cell morphology were negative for HNK-1 (Fig. 3e). After 48 hr in culture, this difference was still present and even more obvious. In early-explant cultures, most of the neural crest cells stained positive for HNK-1 (Fig. 3b), while this percentage was much lower in the late explants (Fig. 3f). Early- and late-explant cultures were also stained for HHF35 to display actin expression. On average, 20% of the neural crest cells in the early explants showed actin staining (Fig. 3c and d). This is in contrast to the late explants, in which up to 90% of the cells that showed neural crest cell characteristics stained positive for actin (Fig. 3g and h). Heterochronic Transplantations The cardiac neural crest of an HH12 quail donor was transplanted into an HH10 chick embryo host in which the cardiac neural crest had been ablated. At HH30 the late-migrating cardiac neural crest cells were stained for the quail marker QCPN, and were shown to be located in the vessel walls of the third, fourth, and sixth pharyngeal arch arteries (Fig. 4a). The cells were found in the proximal and distal parts of the pharyngeal arch arteries. Quail cells were also noted in the nerve bundles surrounding the vessels (Fig. 4a). Many quail cells were observed in the condensed mesenchyme and the prongs of the aorticopulmonary septum at HH30 (Fig. 4b), as well as in the outﬂow tract cushions (Fig. 4c). 1014 BOOT ET AL. Figures 2– 4. CARDIAC NEURAL CREST SUBPOPULATIONS 1015 Differences in distribution patterns between early- and late-migrating cardiac neural crest cells were clearly visible by HH30. The ﬁrst difference noted was that a high number of early-migrating cardiac neural crest cells were located in the condensed mesenchyme and the prongs of the aorticopulmonary septum, while the late-migrating cells were nearly absent in the septum. Removal of the late-migrating cardiac neural crest resulted in normal outﬂow tract septation, which suggests that the late-migrating cardiac neural crest cells are dispensable in the septation process, and the early-migrating cardiac neural crest cells play an essential role in outﬂow tract septation. The notion that cardiac neural crest cells can be divided into two subpopulations may explain the results found in previous ablation experiments. These experiments showed that abnormal patterning of the aortic arch arteries was found in combination with normal cardiac outﬂow tract development (Kirby et al., 1997), as well as in embryos with a combination of persistent truncus arteriosus and relatively normal aortic arches (Nishibatake et al., 1987). This demonstrates that pharyngeal arterial arch and outﬂow tract morphogenesis must be considered as two separate mechanisms. Our ﬁndings indicate that the cardiac neural crest population should be separated into two groups with speciﬁc migratory behavior. The second interesting observation was that the earlymigrating cardiac neural crest cells were scattered over the proximal and distal parts of the pharyngeal arch arteries (with the exception of the dorsal aortae). This is in contrast to the late-migrating cardiac neural crest cells, which were concentrated in the proximal, actin-negative area of the pharyngeal arch arteries, close to the heart. The periendothelial actin-staining pattern shifts at around HH27–HH31 from the proximal part of the pharyngeal arch arteries to the distal part of the vessels (Bergwerff et al., 1996). Subsequently, at around HH30, elastin expression is initiated at the proximal part of the pharyngeal arch arteries, followed by the lamellar actin expression that is also initiated at the proximal part of the pharyngeal arteries. The borderline between periendothelial actin-negative and -positive parts of the arteries nicely coincides with the boundary describing the area of the late-migrating cardiac neural crest cells. The contribution of the late-migrating cardiac neural crest cells to the proximal part of the pharyngeal arch arteries may indicate a role for these cells in remodeling of the pharyngeal arch arteries. This role is supported by the results from the embryos in which the late-migrating cardiac neural crest was removed. These embryos showed hypoplasia, as well as abnormal vessel wall integrity of the pharyngeal arch arteries (especially in the proximal part of the arteries close to the heart). The third difference between the early- and late-migrating populations was observed in the wall of the aorta and the pulmonary trunk just downstream of valve level. The early-migrating cardiac neural crest cells were found at the medial side of these vessels, as described previously in quail-chick chimeras (Waldo et al., 1998). Surprisingly, the late-migrating cardiac neural crest cells were also contributing to the peripheral, ventral side of the base of the aorta, extending with a few cells into the outﬂow tract cushions. The contribution of cardiac neural crest cells to both the medial and peripheral sides of the aorta just above valve level has been described for Cx43-lacZ transgenic mice, but has not been reported in chick embryos (Waldo et al., 1999). Studies with various neural crest reporter mice (Cx43lacZ (Waldo et al., 1999), Pax3-Cre/LacZ (Epstein et al., 2000), and Wnt1-Cre/R26R (Jiang et al., 2000)) have shown differences in neural crest distribution patterns. For example, neural crest cells were observed in the ectoderm covering the pharyngeal arches in the Cx43-lacZ Fig. 2. The experimental procedures used to discriminate between early- and late-migrating cardiac neural crest cells. a: Representation of the experimental procedure: injection of lacZ-retrovirus into neural tube lumen at HH10, ablation of the cardiac neural crest at HH11, and the position of the early-migrating cardiac neural crest cells in the heart and vessels at HH31. The labeled cells are located in the proximal and distal parts of the pharyngeal arch arteries and the aorticopulmonary septum. b: Representation of the experimental procedure: injection of lacZ-retrovirus into neural tube lumen at HH12 and the position of the latemigrating cardiac neural crest cells in the heart and vessels at HH30. Note that labeled cells are nearly absent in the condensed mesenchyme and prongs of the aorticopulmonary septum (yellow structure), and in the distal part of the pharyngeal arch arteries. Abbreviations: III, IV, VI, third, fourth, and sixth pharyngeal arch arteries; L, left; R, right. Fig. 3. In vitro culture. a– d: Early (HH10) neural tube explants. e– h: Late (HH12) neural tube explants. a: After 24 hr almost all neural crest cells (NC) surrounding the early neural tube explant (E) stain positive for HNK-1. b: The 48-hr culture of an early explant shows a high number of HNK-1 positive cells. c: The early explant displays a low number of actin-expressing cells after 48 hr in culture. d: The actin-expressing cells of part c shown in higher magniﬁcation. e: Only a few neural crest (NC) cells are positive for HNK-1; however, a high number of cells with the neural crest morphology are negative for HNK-1 in this late-explant culture after 24 hr (arrows). f: The number of HNK-1 expressing cells in the late-explant culture is low after 48 hr g: Staining for actin displayed high numbers of smooth muscle cells in the 48-hr late-explant cultures. h: The actin-expressing cells in the late explants shown in higher magniﬁcation. Abbreviations: E, explant; N, neuroepithelial cells; NC, neural crest cells. Bar ⫽ (d and h) 250 m, and (a– c, e– g) 500 m. Fig. 4. Heterochronic transplantations. A quail HH12 late cardiac neural crest was transplanted into an HH10 chick embryo and survived to HH30. Late-migrating quail cardiac neural crest cells are observed in (a) the vessel walls of the pharyngeal arch arteries (arrows) and nerve tissue (arrowheads), as well as in (b) the condensed mesenchyme (CM) of the aorticopulmonary septum (arrows) and (c) the outﬂow tract cushions (arrows). Bar ⫽ 150 m. Abbreviations: Ao, aorta; P, pulmonary trunk; III, IV, VI, third, fourth, and sixth pharyngeal arch arteries; L, left; R, right. DISCUSSION The cardiac neural crest has been considered to be a contiguous group of cells with full potential to contribute to the various parts of the developing heart and pharyngeal arch arteries. As described in the present study, the cardiac neural crest can be temporally separated into at least two subpopulations. The early-migrating cardiac neural crest cells leave the dorsal midline of the neural tube at about HH10, and the late-migrating population emigrates from the dorsal side of the neural tube about 6 hr later (HH12). Differences in Fate and Function of Early- and Late-Migrating Cardiac Neural Crest Cells 1016 BOOT ET AL. mice, but they were not seen in the Wnt1-Cre/R26R neural crest reporter mice. In Wnt1-Cre/R26R mice, an asymmetrical labeling pattern of neural crest cells was seen in the pharyngeal arteries at E12.0, which has not been described in other neural crest cell reporter mice. The differences in neural crest staining in these mouse models may be explained by a temporal-spatial expression of the neural crest-related genes Wnt1, Cx43, and Pax3. We hypothesize that timing differences in gene expression could separate the neural crest cell population in early- and late-migrating neural crest cells, which target different positions in the developing cardiovascular system. Restricted Developmental Fate of LateMigrating Cardiac Neural Crest Cells At HH30 the early-migrating cardiac neural crest cells contributed to the outﬂow tract, and both the proximal and distal parts of the pharyngeal arch arteries, whereas the late-migrating cardiac neural crest cells appeared to be restricted to the proximal part of the pharyngeal arch arteries. Differences in patterning of the early- and late-migrating cardiac neural crest cells may be due to intrinsic differences. Studies by Newgreen and Thiery (1980) showed differences in ﬁbronectin expression between early- and late-migrating cranial and sacral crest cells. The early crest cells were able to synthesize ﬁbronectin, while the later population was not, which suggests that the neural crest cells became segregated over time. In our in vitro experiments (this study) on the intrinsic differences between early- and late-migrating cardiac neural crest cells, we found that the number of late-migrating cardiac neural crest cells that differentiated into smooth muscle cells was higher than that found in early-neural tube explant cultures. By combining our ﬁndings that 1) latemigrating cardiac neural crest cells predominantly differentiate into actin-expressing cells, and 2) late-migrating cardiac neural crest cells migrate to a location within the pharyngeal arch arteries where elastogenesis and lamellar actin expression are initiated, we could hypothesize that late-migrating cardiac neural crest cells are involved in the induction of lamellar actin expression. The in vitro results suggested that the early- and late-migrating cardiac neural crest cells have intrinsic differences. However, we questioned whether these differences should be considered as restrictions in developmental potential, as described for the trunk neural crest cells. Trunk neural crest cultures and studies of DiI-labeled neural crest cells (Artinger and Bronner-Fraser, 1992) have shown that early-migrating trunk neural crest cells differentiate into many derivatives, including pigment cells, neurons, and adrenergic cells, while late-migrating cells do not differentiate into adrenergic cells. This suggests that late-migrating trunk neural crest cells have a restricted developmental potential. In our heterochronic transplantations at the cardiac neural crest level, we showed that the late-migrating cardiac neural crest cells are not fully restricted in their developmental potential. The late-migrating cardiac neural crest cells migrated in an early (HH10) environment, in the absence of early-migrating cardiac neural crest cells, to the condensed mesenchyme and prongs of the aorticopulmonary septum, indicating the strong inﬂuences of the early-migrating cardiac neural crest cells and the tissues surrounding the migrating cardiac neural crest cells. The effects of regional differences on the migratory behavior of individual neural crest cells and development of cell depletion areas along the rostrocaudal axis in the head have been demonstrated by in ovo time-lapse microscopy (Kulesa and Fraser, 1998). Furthermore, the importance of the spatial distribution of inﬂuential factors and genes (e.g., Msx2 and Bmp4) has been shown in the neural crest migration environment (Graham et al., 1994). Tbx1 is another gene that has been described as an environmental gene that affects neural crest migration. Tbx1 is expressed in the pharyngeal endoderm surrounding the pharyngeal arch arteries, and the Tbx1–/– mutant displays early growth defects of the pharyngeal arches in combination with disturbed neural crest cell migration (Vitelli et al., 2002a,b). We postulate that the migration pathways of the latemigrating cardiac neural crest cells are affected by the presence of the early-migrating cardiac neural crest cells. The early-migrating cardiac neural crest cells may send inhibitory or guiding signals to the late-migrating cardiac neural crest cell population, thereby either restraining them in the pharyngeal arch arteries, or preventing them from migrating into the already established aorticopulmonary septum and diverting them to the lateral side of the aorta and the outﬂow tract cushions. The hypothesis that the early-migrating cells have guiding functions is supported by studies on the mesencephalic crest (Baker et al., 1997), in which the late-migrating cranial neural crest cells form neurons in cranial ganglia but not in cartilage of the jaw. By ablating the early-migrating cells and heterochronic (late-to-early) quail-to-chick neural crest grafting, it has been shown that the late-migrating cells are not lineage-restricted and are able to contribute to the jaw skeleton after ablation of the early-migrating cells. This suggests that the absence of late-migrating cells in the jaws may be regulated by the presence of earlier-migrating cells. The notion that neural crest cell groups inﬂuence each other was also suggested by an analysis of the hindbrain, in which cell– cell contact was shown in streams of neural crest cells that were bound for different branchial arches (Kulesa and Fraser, 2000). Our retroviral and ablation studies showed that the differences in early- and late-migrating cardiac neural crest cell patterning were not obvious before HH30, indicating that at around HH30 there is probably another regulatory mechanism in the area of the pharyngeal arch arteries and aorticopulmonary septum. It is conceivable that the late-migrating cells receive their information in the pharyngeal arches and close to the aorticopulmonary septum. In addition to environmental effects and signaling by cell populations, the genes expressed by the cardiac neural crest cells themselves play an important role in the differentiation and migration differences between the earlyand late-migrating cardiac neural crest cells. The neural crest cells are probably separated into subpopulations by the expression of different neural crest-related genes, such as Pax3, Wnt1, BMP2, and BMP4. The idea that cardiac neural crest subpopulations have their own speciﬁc gene expression patterns is supported by the differences in neural crest distribution shown in reporter mice (Waldo et al., 1999; Epstein et al., 2000; Jiang et al., 2000). Previous ﬁndings in Xenopus (Borchers et al., 2001) like- CARDIAC NEURAL CREST SUBPOPULATIONS wise showed expression patterns of Xcad-11, AP-2, twist, and Snail that only partly overlapped in the migrating cranial neural crest cell population. Borchers et al. (2001) also postulated the existence of time-induced subpopulations in neural crest migration. Finally, we have to keep in mind the extensive remodeling of the outﬂow tract. This involves elongation from the secondary heart ﬁeld (Waldo et al., 2001), and several cell populations that deﬁne the tissue boundaries in this complex region (Noden et al., 1995). The 6-hr time difference between the time points at which the early- and late-migrating cardiac neural crest cells leave the neural tube might result in the late-migrating cardiac neural crest cells lagging behind. This would imply that earlymigrating cells reach the condensed mesenchyme and prongs of aorticopulmonary septum, while most of the late-migrating cells stay anterior to the elongating conotruncus. Implications of Time-Induced Subpopulations of Cardiac Neural Crest Cells The exact timing of the migratory cardiac neural crest cell subpopulations is essential for normal cardiovascular development. The ﬁnding that aorticopulmonary septum formation depends on early-migrating (but not on latemigrating) cardiac neural crest cells in chick embryos may indicate that the focus of the development of outﬂow tract septation malformations, such as persistent truncus arteriosus, should be directed toward the early stages of cardiac neural crest cell migration. By narrowing the developmental range in which congenital heart diseases arise, we may in the near future be able to exclude a number of supposed candidate genes involved in neural crest-related congenital heart diseases, such as DiGeorge syndrome and velocardiofacial syndrome. ACKNOWLEDGMENTS We thank Kim van der Heiden for technical assistance. The artwork and photographic contributions by Ron Slagter, Bas Blankenvoort, and Jan Lens were indispensable. The retroviral construct was kindly provided by Dr. Takashi Mikawa. LITERATURE CITED Abo T, Balch CM. 1981. A differentiation antigen of human NK and K cells identiﬁed by a monoclonal antibody (HNK-1). J Immunol 127: 1024 –1029. Artinger KB, Bronner-Fraser M. 1992. Partial restriction in the developmental potential of late emigrating avian neural crest cells. Dev Biol 149:149 –157. Baker CVH, Bronner-Fraser M, Le Douarin NM, Teillet MA. 1997. Early- and late-migrating cranial neural crest cell populations have equivalent developmental potential in vivo. Development 124: 3077–3087. Bergwerff M, DeRuiter MC, Poelmann RE, Gittenberger-de Groot AC. 1996. Onset of elastogenesis and downregulation of smooth muscle actin as distinguishing phenomena in artery differentiation in the chick embryo. Anat Embryol 194:545–557. Bergwerff M, Verberne ME, DeRuiter MC, Poelmann RE, Gittenberger-de Groot AC. 1998. Neural crest cell contribution to the developing circulatory system. Implications for vascular morphology? Circ Res 82:221–231. Bockman DE, Redmond ME, Kirby ML. 1989. Alteration of early vascular development after ablation of cranial neural crest. Anat Rec 225:209 –217. Boot MJ, Gittenberger-de Groot AC, van Iperen L, Poelmann RE. 2003. The myth of ventrally emigrating neural tube (VENT) cells 1017 and their contribution to the developing cardiovascular system. Anat Embryol (Berl) 206:327–333. Borchers A, David R, Wedlich D. 2001. Xenopus cadherin-11 restrains cranial neural crest migration and inﬂuences neural crest speciﬁcation. Development 128:3049 –3060. Epstein JA, Li J, Lang D, Chen F, Brown CB, Jin F, Lu MM, Thomas M, Liu E, Wessels A, Lo CW. 2000. Migration of cardiac neural crest cells in Splotch embryos. Development 127:1869 –1878. Garg V, Yamagishi C, Hu T, Kathiriya IS, Yamagishi H, Srivastava D. 2001. Tbx1, a DiGeorge syndrome candidate gene, is regulated by sonic hedgehog during pharyngeal arch development. Dev Biol 235: 62–73. Goldmuntz E, Emanuel BS. 1997. Genetic disorders of cardiac morphogenesis. The DiGeorge and velocardiofacial syndromes. Circ Res 80:437– 443. Graham A, Francis-West P, Brickell P, Lumsden A. 1994. The signalling molecule BMP4 mediates apoptosis in the rhombencephalic neural crest. Nature 372:684 – 686. Hamburger V, Hamilton HL. 1951. A series of normal stages in the development of the chick embryo. J Morphol 88:49 –92. Jiang X, Rowitch DH, Soriano P, McMahon AP, Sucov HM. 2000. Fate of the mammalian cardiac neural crest. Development 127:1607– 1616. Kirby ML, Gale TF, Stewart DE. 1983. Neural crest cells contribute to normal aorticopulmonary septation. Science 220:1059 –1061. Kirby ML, Turnage KL, Hays BM. 1985. Characterization of conotruncal malformations following ablation of “cardiac” neural crest. Anat Rec 213:87–93. Kirby ML, Waldo KL. 1990. Role of neural crest in congenital heart disease. Circulation 82:332–340. Kirby ML. 1993. Cellular and molecular contributions of the cardiac neural crest to cardiovascular development. Trends Cardiovasc Med 3:18 –23. Kirby ML, Waldo KL. 1995. Neural crest and cardiovascular patterning. Circ Res 77:211–215. Kirby ML, Hunt P, Wallis K, Thorogood P. 1997. Abnormal patterning of the aortic arch arteries does not evoke cardiac malformations. Dev Dyn 208:34 – 47. Kochilas L, Merscher-Gomez S, Lu MM, Potluri V, Liao J, Kucherlapati R, Morrow B, Epstein JA. 2002. The role of neural crest during cardiac development in a mouse model of DiGeorge syndrome. Dev Biol 251:157–166. Kulesa PM, Fraser SE. 1998. Neural crest cell dynamics revealed by time-lapse video microscopy of whole embryo chick explant cultures. Dev Biol 204:327–344. Kulesa PM, Fraser SE. 2000. In ovo time-lapse analysis of chick hindbrain neural crest cell migration shows cell interactions during migration to the branchial arches. Development 127:1161–1172. Le Lièvre Cs, Le Douarin NM. 1975. Mesenchymal derivatives of the neural crest: analysis of chimaeric quail and chick embryos. J Embryol Exp Morphol 34:125–154. Mikawa T, Fischman DA, Dougherty JP, Brown AMC. 1991. In vivo analysis of a new lacZ retrovirus vector suitable for cell lineage marking in avian and other species. Exp Cell Res 195:516 –523. Newgreen DF, Thiery JP. 1980. Fibronectin in early avian embryos: synthesis and distribution along the migration pathways of neural crest cells. Cell Tissue Res 211:269 –291. Nishibatake M, Kirby ML, van Mierop LHS. 1987. Pathogenesis of persistent truncus arteriosus and dextroposed aorta in the chick embryo after neural crest ablation. Circulation 75:255–264. Noden DM, Poelmann RE, Gittenberger-de Groot AC. 1995. Cell origins and tissue boundaries during outﬂow tract development. Trends Cardiovasc Med 5:69 –75. Poelmann RE, Mikawa T, Gittenberger-de Groot AC. 1998. Neural crest cells in outﬂow tract septation of the embryonic chicken heart: differentiation and apoptosis. Dev Dyn 212:373–384. Poelmann RE, Gittenberger-de Groot AC. 1999. A subpopulation of apoptosis-prone cardiac neural crest cells targets to the venous pole: multiple functions in heart development? Dev Biol 207:271–286. Poelmann RE, Molin D, Wisse LJ, Gittenberger-de Groot AC. 2000. Apoptosis in cardiac development. Cell Tissue Res 301:43–52. 1018 BOOT ET AL. Schiafﬁno S, Dallapiccola B, Di Lisi R. 1999. Molecular genetics of congenital heart disease. A problem of faulty septation. Circ Res 84:247–249. Thomas T, Kurihara H, Yamagishi H, Kurihara Y, Yazaki Y, Olson EN, Srivastava D. 1998. A signaling cascade involving endothelin-1, dHAND and msx1 regulates development of neural-crestderived branchial arch mesenchyme. Development 125:3005– 3014. Tosney KW. 1982. The segregation and early migration of cranial neural crest cells in the avian embryo. Dev Biol 89:13–24. Tsukada T, Tippens D, Gordon D, Ross R, Gown AM. 1987. HHF35, a muscle-actin-speciﬁc monoclonal antibody. I. Immunocytochemical and biochemical characterization. Am J Pathol 126:51– 60. Verberne ME, Gittenberger-de Groot AC, Poelmann RE. 1998. Lineage and development of the parasympathetic nervous system of the embryonic chicken heart. Anat Embryol 198:171–184. Verberne ME, Gittenberger-de Groot AC, VanIperen L, Poelmann RE. 2000. Distribution of different regions of cardiac neural crest in the extrinsic and the intrinsic cardiac nervous system. Dev Dyn 217: 191–204. Vitelli F, Morishima M, Taddei I, Lindsay EA, Baldini A. 2002a. Tbx1 mutation causes multiple cardiovascular defects and disrupts neural crest and cranial nerve migratory pathways. Hum Mol Genet 11:915–922. Vitelli F, Taddei I, Morishima M, Meyers EN, Lindsay EA, Baldini A. 2002b. A genetic link between Tbx1 and ﬁbroblast growth factor signaling. Development 129:4605– 4611. Waldo K, Miyagawa-Tomita S, Kumiski D, Kirby ML. 1998. Cardiac neural crest cells provide new insight into septation of the cardiac outﬂow tract: aortic sac to ventricular septal closure. Dev Biol 196:129 –144. Waldo KL, Lo CW, Kirby ML. 1999. Connexin 43 expression reﬂects neural crest patterns during cardiovascular development. Dev Biol 208:307–323. Waldo KL, Kumiski DH, Wallis KT, Stadt HA, Hutson MR, Platt DH, Kirby ML. 2001. Conotruncal myocardium arises from a secondary heart ﬁeld. Development 128:3179 –3188.