Cardiac outflow tract: A review of some ... - Wiley Online Library

56 downloads 0 Views 138KB Size Report
Aug 4, 2006 - reported by Keith (1909). The conotruncus comprises col- lectively two myocardial subsegments, the conus and the truncus, although the exact ...
THE ANATOMICAL RECORD PART A 288A:936–943 (2006)

Cardiac Outflow Tract: A Review of Some Embryogenetic Aspects of the Conotruncal Region of the Heart ANGELO RESTIVO, GERARDO PIACENTINI, SILVIA PLACIDI, CLAUDIA SAFFIRIO, AND BRUNO MARINO* Pediatric Cardiology, Department of Pediatrics, University of Rome ‘‘La Sapienza,’’ Rome, Italy

ABSTRACT A review concerning some embryogenetic aspects of the cardiac outflow tract is presented. Two main topics are discussed: the truncal septation and the secondary heart field. In the context of the septation of the truncus arteriosus, the development of the arterial valves is largely discussed, particularly in reference to the sinuses of Valsalva. Emphasis is also given to the fate of the external myocardial wall of the truncus arteriosus, as this primordial myocardial surface disappears later in the development. Molecular genetics data concerning Sox4 and NF-Atc transcription factors are correlated in the present review with rare forms of truncus malformations encountered in human pathology. The roles exerted by the secondary heart field and the neural crest on the development and growth of the conotruncal musculature are largely discussed. Reported experimental ablations of both secondary heart field and neural crest, showed conotruncal defects such as persistent truncus arteriosus, tetralogy of Fallot, and double-outlet right ventricle, which were considered as the result of a short outflow tract causing, ultimately, a lack of conotruncal rotation. In this regard, some morphologic correlations are carried out, in the present review, between these experimental animal models and human malformations, and it is thought that this sort of conotruncal defects cannot be explained always in terms of conotruncal hypoplasia. Finally, influence of Pitx2c, a left-right laterality signaling gene, on the modulation of the conotruncal rotation, as most recently reported, is emphasized in terms of very likely multifactorial contributions in the embryogenesis of the conotruncal region of the heart. Anat Rec Part A, 288A:936–943, 2006. Ó 2006 Wiley-Liss, Inc.

Key words: outflow tract; conotruncus; truncal septation; secondary heart field; neural crest

In this review, we wish to discuss some aspects concerning the embryonic development of the cardiac outflow tract (OFT), well known as ‘‘conotruncus.’’ One of the earliest appearances in the literature of the term ‘‘conus,’’ known also as ‘‘infundibulum,’’ is probably that reported by Keith (1909). The conotruncus comprises collectively two myocardial subsegments, the conus and the truncus, although the exact borderline between the two is almost unassessable in mammals. Similarly, the developing two main truncal cushions and the underlying two conal cushions are perfectly aligned and no line of demarcation between the two is identifiable in mammals (Van Mierop and Patterson, 1980); in contrast, indentaÓ 2006 WILEY-LISS, INC.

tions seem to be present between truncal and conal cushions in avians (De la Cruz et al., 1977). In spite of this almost unidentifiable anatomical borderline in some species, the distinction between the two portions of the *Correspondence to: Bruno Marino, Department of Pediatrics, University of Rome ‘‘La Sapienza,’’ V. le Regina Elena 324, 00161 Rome, Italy. Fax: 0039649970356. E-mail: [email protected] Received 6 March 2006; Accepted 12 June 2006 DOI 10.1002/ar.a.20367 Published online 4 August 2006 in Wiley InterScience (www.interscience.wiley.com).

CARDIAC OUTFLOW TRACT

937

in the definitive normal heart as subpulmonary infundibulum, whereas it disappears, by myocardial regressive absorption, in the subaortic side (Goor et al., 1972; Pexieder, 1975). In the present review, we focus our attention on two main specific aspects of the conotruncal development, namely, the septation of the truncus arteriosus and related development of the arterial valves and the secondary heart field. As far as truncal septation and development of the arterial valves are concerned, the exact derivative source of the arterial walls of the sinuses of Valsalva is unclear. Similarly, the embryogenetic mechanism allowing the disappearance of the myocardial wall of the truncus arteriosus does not seem to be entirely clear. The secondary heart field (SHF) is also the object of extensive discussion in the present review. This specific area of the ventral pharyngeal mesoderm, identified in most recent years (Kelly et al., 2001; Mjaadvedt et al., 2001; Waldo et al., 2001), is considered as the precardiac splanchnic mesodermic region providing not only myocardial precursor cells, which migrate to the OFT area of the developing cardiac tube, where they build up the conotruncal myocardium, but also smooth muscle cells joining the caudal portion of the aortic sac (Waldo et al., 2005b). The detailed molecular modulation of the epithelialmesenchymal transformation, leading to the development, growth, and fusion of the conotruncal endocardial cushions, is beyond the scope of the present review.

TRUNCAL SEPTATION: DEVELOPMENT OF THE ARTERIAL VALVES

Fig. 1. The arterial pole includes the vascular aortic sac, from which the bilateral branchial arch arteries arise, and the cardiac conotruncus. The truncal cushions are shown horizontally along the truncal circumference, whereas the conal cushions are depicted in a longitudinal view.

OFT is crucial since the development of these two subsegments leads to different anatomical structures in the definitive normal heart: arterial valves and subpulmonary infundibulum. A graphic illustration, showing the arterial pole of the embryonic cardiac tube and the truncal and conal cushions, is reported in Figure 1. The truncus arteriosus, the most distal portion of the developing cardiac outflow tract, bordering on the overlying aortic sac, is the short segment that, once septated, allows the division of the common outflow orifice into two separate arterial valves orifices. This septation is accomplished by the developing truncal endocardial cushions, which are also crucial for the development of the arterial valves leaflets. It is essential to point out, as emphasized by Van Mierop et al. (1978), that ‘‘truncus’’ in embryological terms should not be extended to the common arterial trunk, which is an aortic sac derivative from which the suprasemilunar ascending aorta and pulmonary trunk arteries develop. The conus is that portion of the outflow tract proximal to the ventricles, which, after proper septation and rotational remodeling, persists

As already mentioned, the truncus arteriosus, the myocardial segment of the OFT interposed between the conus and the aortic sac, undergoes complex developmental processes leading to the formation of the arterial valves. At an earlier stage of development, the external wall of the truncal myocardium creates some sort of rim known as myocardial cuff, which appears to cover the root of the aortic sac (Thompson and Fitzharris, 1979a). At the same time, two main opposing dextrosuperior and sinistroinferior truncal endocardial cushions appear. Occupying respectively a dorsal and a ventral oblique position, these cushions extend from the junction between the aortic sac and the truncus arteriosus down to the beginning of the conus, where they align with the dextrodorsal and sinistroventral conal cushions, respectively (Van Mierop and Patterson, 1980). The truncal cushions, once fused, form the truncal septum. The mesenchymal truncal septum undergoes a complex differentiation process leading to the formation of the posterior right and left pulmonary valve cusps and of the anterior right coronary and left coronary aortic valve cusps. Two additional intercalated truncal endocardial swellings also appear to occupy a parietal position on the right and on the left side of the truncus arteriosus. Following the normal counterclockwise conotruncal rotation, looking from the ventricular side, the right intercalated cushion becomes the posterior noncoronary aortic valve cusp and the left intercalated cushion becomes the anterior pulmonary valve cusp. Subsequently, the myocardial cuff retracts downward following the caudal translocation of the aortic sac and the whole conotruncus itself moves caudally; this concept of myocardial cuff retraction was well

938

RESTIVO ET AL.

established by Thompson and Fitzharris (1979a, 1979b). More specifically, these authors considered the expansion of tissue from the aortic sac down along the outflow tract as the result of a physical translocation due to biophysical forces; indeed, no cellular migratory activity has ever been shown in this regard. Nevertheless, this mechanism of myocardial regression is not entirely clear yet. Indeed, we wonder how this downward translocation of the outflow tract can cause the disappearance of the external myocardial wall of the developing arterial valves. In this respect, Ya et al. (1998a) and more recently Anderson et al. (2003) reemphasized the concept originally advanced by Patten (1964) and Arguello et al. (1978) according to which, in the definitive heart, the smooth muscle tissue of the sinuses of Valsalva and even the intrapericardial portions of the great arteries are the result of a transdifferentiation of the myocardial mantel of the embryonic truncus arteriosus. As far as this specific aspect is concerned, we question how the already differentiated cardiomyocytes dedifferentiate switching off from a myosin heavy chain (MHC) to a new a-smooth muscle actin (a-SMA) genotype program and phenotype sequel. Furthermore, Ya et al. (1998a) emphasize the fact that no apoptosis was found at the truncal level; consequently, they excluded the fact that the truncal cardiomyocytes were replaced by a local invasion of mesodermic mesenchymal cells differentiating subsequently into smooth muscle cells. In contrast, Waldo et al. (2005b) show that SHF pharyngeal mesoderm provides smooth muscle precursor cells to the caudal portion of the aortic sac. Indeed, according to these latest findings, it is assumed that the caudal elongation of the aortic sac, concomitantly with the downward retraction of the truncus arteriosus, is the one that allows the development of the intrapericardial portions of the great arteries and of the arterial walls of the sinuses of Valsalva. However, it does not seem to be established yet, to the best of our knowledge, how the aortic sac incorporates the developing sinuses of Valsalva if the entire truncus is moving downward. Now, in this regard, we are tempted to advance a speculation. Indeed, the proximal conus absorption pulling down the distal conus and the truncus, as indicated by several authors (Goor et al., 1972; Pexieder, 1975; Watanabe et al., 2001), might produce, we think, some sort of sliding of the myocardial cuff down along the cardiac jelly, thereby allowing the aortic sac to envelop the intercalated truncal swellings and ultimately the developing sinuses of Valsalva. In this respect, it is also important to emphasize, as previously mentioned, that since no apoptosis was identified along the truncal myocardial wall (Ya et al., 1998a), as a consequence, the myocardial retraction is not to be referred to as a programmed cell death. Specifically, as far as the development of the sinuses of Valsalva is concerned, Anderson et al. (2003) emphasize the fact that some sort of cavitation is supposed to occur within the truncal and the intercalated endocardial cushions. According to these authors, indeed, this sort of excavation produces a central luminal part of each cushion to form the arterial valvar leaflets, whereas the peripheral part becomes arterialized to form the wall of the supporting valvar sinuses. More specifically, in the case of the intercalated cushions, Anderson et al. (2003), in support of the transdifferentiation hypothesis of the truncal myocardium to smooth

muscle tissue, hypothesize that the parietal surfaces of the sinuses of Valsalva related to the intercalated cushions are due to the arterialization of the truncal myocardial mantel, whereas the smooth muscle surface of the other four central cusps is supposed to be the result of differentiation of the truncal cushion mesenchyme. In contrast, considering the myocardial cuff retraction (Thompson and Fitzharris, 1979a) and the caudal elongation of the aortic sac supplied by SHF smooth muscle precursor cells (Waldo et al., 2005b) and in considering further that such a precursor cells might form the smooth muscle wall of the sinuses of Valsalva, a question arises, we think, about the potentially different derivative source of the arterial walls of these sinuses. Indeed, while the arterial surface walling the intercalated truncal swellings would be most probably of aortic sac origin, as previously emphasized (Waldo et al., 2005b), on the other hand, the arterial surface related to the two main septal truncal cushions should be the result of a-SMA differentiation of the local truncal endocardial cushion mesenchymal cells themselves, as Anderson et al. (2003) seem to suggest. In considering the more advanced stage of maturation of the truncal cushions toward the arterial valves formation, Sox4 and NF-Atc transcription factors were found to be mainly involved in this developmental phase. Indeed, Schilham et al. (1996) and Ya et al. (1998b) developed mouse studies where Sox4 / homozygous experimental models caused anomalous development of the truncal ridges leading either to dysplastic arterial valves or even to a common truncal valve (truncus malformation). Thus, these investigations show that such a Sox4 transcription factor somehow regulates the normal development and fusion of the truncal endocardial cushions. Similarly, Ranger et al. (1998) reported another mouse experimental study where targeted disruption of the NF-Atc gene produced absence of both arterial valves. As these authors point out, Sox4 probably plays an earlier role compared to NF-Atc in the development of the arterial valves. Now, we wish to analyze and discuss these two studies collectively and to make some considerations in this regard. If we compare these studies, we can appreciate the fact that in Sox4 / model the truncal ridges failed to fuse producing a common truncal valve (truncus malformation), whereas in the NF-Atc / homozygous model, the intrapericardial portions of the great arteries were normally formed and septated but no valvar leaflets were identifiable in the site of the arterial valve orifices. These data seem controversial at first glance in the sense that, in the NF-Atc / model, the fact that two arterial valve orifices are present without valvar leaflet tissue could be interpreted as if the dextrosuperior truncal cushion had been fusing with the sinistroinferior truncal cushion, producing then a normal truncal septation. If so, two separate arterial valve orifices would be formed without fibroblastic maturation of the truncal septum components, normally producing the anterior right and left coronary aortic valve cusps and the posterior right and left pulmonary valve cusps, and also without any developmental evolution of the intercalated truncal swellings, which normally differentiate and produce the anterior pulmonary valve cusp and the posterior noncoronary aortic valve cusp. Looking at the Sox4 / model, the common truncal valve means that no truncal ridges fusion has occurred but, despite that, the unfused truncal cushions and the intercalated truncal swellings were able to mature and

939

CARDIAC OUTFLOW TRACT

produce truncal valvar leaflets. Thus, the controversy seems to be the fact that a fusion of the truncal cushions would imply a more advanced sort of developmental stage with a more or less normal valvar leaflets development to be expected. On the contrary, in the Sox4 / model, the failure of truncal cushion fusion coexists with the full development of the truncal valve leaflets. Having said that, however, we are not entirely sure on whether or not the NF-Atc / model reported by Ranger et al. (1998) could really be interpreted as if the truncal septum was actually formed. Indeed, we could speculate that in such a model, a normal and intact A-P septum, dividing the aortic sac into separate ascending aorta and pulmonary trunk, might coexist with a lack of development of the truncal endocardial cushions. Incidentally, extremely rare cases of persistent truncus arteriosus (PTA) malformation can show a coexistence of two separate and fully formed great arteries with a common truncal valve overriding the ventricles, with or without infundibular ventricular septal defect (Rosenquist et al., 1976; Carr et al., 1979). In this respect, it is important to point out that the essence of the PTA malformation is the absence of the truncal septum, leading postnatally to a common truncal valve, due to maldevelopment and lack of fusion of the truncal cushions, a concept originally expressed by Van Mierop et al. (1978) and reemphasized by Carr et al. (1979). Thus, although in the great majority of cases of PTA, there is also a lack of A-P septum and of conal septum, this combination is not absolute and there can be some variability in this regard, up to the extremely rare cases as the ones reported by Rosenquist et al. (1976) and Carr et al. (1979). In considering the postnatal morphology, if we look at the most common forms of truncus malformation, with typical biventricular common truncal valve, and excluding the cases where the truncal valve arises exclusively from the right ventricle or, less commonly, from the left ventricle, we can appreciate that the infundibular septum cannot be entirely absent and that, indeed, some rudimentary conal septum can be identified arching above the membranous septum, producing a tricuspid-truncal valve fibrous discontinuity and making up some sort of muscle band, which creates an arch of continuity between the posterior limb of the septomarginal trabeculation and the ventriculoinfundibular fold (Baker and Anderson, 1981). Occasionally, a complete muscular subtruncal conus can be found in the cases where the truncal valve is entirely related to the right ventricle. If we move to the A-P septum, the truncus types 2, 3, and 4 represent a complete failure of A-P septation, whereas in truncus type 1, there is only a partial A-P septation deficiency being the pulmonary trunk present. In this specific context, concerning the anatomic variations in the pattern of origin of the pulmonary and aortic tributaries from the ascending arterial trunk, it is important to refer to the study of Collett and Edwards (1949), who provided, to the best of our knowledge, the first clear-cut classification of truncus malformation. According to this classification, in truncus type 1, a short pulmonary trunk emerges from the left lateral wall of the common arterial trunk and divides into right and left pulmonary branches. In type 2, the pulmonary arteries arise either independently or via a common orifice from the posterior wall of the common arterial trunk. In type 3, only one pulmonary artery (right or

left) is present, and one lung is supplied by either a bronchial artery or an aberrant pulmonary artery arising from the aortic arch. In type 4, no true pulmonary arteries are present, and the lungs are supplied by arteries arising from either the aortic arch or the descending aorta. In the extremely rare cases of PTA, previously mentioned and morphologically documented by Rosenquist et al. (1976) and by Carr et al. (1979), showing two separate great arteries, even with an intact and fully formed infundibular septum normally aligned and fused with the ventricular septum, but with a common truncal valve orifice, such a truncal orifice showed common truncal valve leaflets tissue. The NF-Atc / experimental model reported by Ranger et al. (1998), on the other hand, shows two separate great arteries with two valve orifices without any valvar leaflet tissue at all. Thus, comparing again the Sox4 / and the NF-Atc / models, it might be possible to speculate that in the first model, epithelial-mesenchymal transformation occurred and that truncal endocardial cushions started to develop anomalously, producing in fact a common truncal valve or, in the less severe cases, two separate dysplastic arterial valves. In the NF-Atc / model, on the other hand, the absence of valvar leaflet tissue would indicate, most probably, a severely deficient or entirely absent epithelial-mesenchymal transformation.

SECONDARY HEART FIELD As originally indicated by De la Cruz et al. (1977) in an experimental study on chicken, a primary heart tube develops first from bilateral primary heart fields located in the lateral plate mesoderm. Secondarily, according to the same author, the atrioventricular (A-V) canal and the sinu-atrial segment, at the venous pole, and the conotruncus, at the arterial pole, are added to the heart tube just in the very initial stage of looping. In recent years, Mjaatvedt et al. (2001), Waldo et al. (2001), and Kelly et al. (2001), in support of the data of De la Cruz et al. (1977), have identified, in separate investigations on chicken and mouse, a secondary or anterior heart field in the ventral pharyngeal mesoderm. These authors showed also that this secondary precardiac mesodermic mesenchyme expressed NKx2.5 and Gata4 transcription factors. According to Waldo et al. (2001), then, NKx2.5and Gata4-committed precardiac cells move from the SHF toward the arterial pole of the primary heart tube. As these cells migrate, they express HNK-1, a cell surface glycoprotein; indeed, the onset of HNK-1 expression in the mesodermal SHF coincides with the caudal translocation of the NKx2.5- and Gata4-committed cells toward the primary heart tube. Once NKx2.5- and Gata4 SHF-committed cells join and incorporate themselves into the outflow tract of the primary heart tube, these cells undergo terminal myocardial differentiation under the induction of the local primary myocardial Bmp2 factor, ultimately allowing normal development of the conotruncal segment. Thus, from these studies, it seems as if the Bmp2 (bone morphogenetic protein) growth factor induces both myocardial commitment and myocardial differentiation. According to Kelly et al. (2001), the anterior heart field is responsible for not only the conotruncus but also the right ventricle in mice. Most recently, Waldo et al. (2005a) found that neural crest (NC) cells modulate the

940

RESTIVO ET AL.

SHF. These authors found that after experimental neural crest ablation in chicken embryos, the SHF precardiac cells failed to join the conotruncus, so the outflow tract could not elongate, remaining short, confirming then this sort of previous observation as reported by Yelbuz et al. (2002). Furthermore, the same authors (Waldo et al., 2005a, 2005b) found also that NC ablation did not interfere with the migration of SHF smooth muscle precursor cells to the caudal portion of the aortic sac; however, the conotruncus being short and unable to expand caudally, as a consequence even the caudal portion of the aortic sac could not expand further downward, failing then to elongate. More specifically, the reason why the NC ablation prevents the migration of the SHF-committed precardiac cells to the conotruncus has to be put in relation, according to these investigators, to the lack of modulation that, normally, the NC cells exert on the Fgf8 growth factor signaling at pharyngeal level. Indeed, according to Waldo et al. (2005a), in normal conditions NC cells seem to keep the Fgf signaling under proper control and consequently a balanced Fgf8/Bmp2 inductive signaling to myocardial differentiation would be maintained. Fgf8 and BMP factors in fact seem to exert antagonistic interactions in the sense that Fgf8 antagonizes BMP signaling to prevent premature and excessive myocardial differentiation (Waldo et al., 2001). Lack of NC cell modulation would cause, in the light of these recent studies (Waldo et al., 2005a) and of previous ones (Farrell et al., 2001), an excess of Fgf8, with consequent overproliferation of SHF precardiac cells, which would remain then confined at the junction between the distal outflow myocardium rim and the most caudal portion of the aortic sac, rather than migrating caudally and differentiating into myocardial cells, allowing ultimately the elongation of the outflow tract as it normally occurs. This point is in contrast with another recent study of Xu et al. (2004), who found that, in normal conditions, the T-box transcription factor TBX-1 expressed in the SHF maintains, through the intermediate Fgf10 induction, a proliferative output of NKx2.5-positive cells (committed cardiomyocytes) from the SHF to the outflow tract. According to these authors, loss of TBX-1 function, which is intimately related to NC cell development, would cause cellular hypoproliferation in the SHF and consequent hypoplasia of the outflow tract myocardium. In contrast, Waldo et al. (2005a) put the NC deficiency and consequent conotruncal maldevelopment in relation to SHF cellular hyperproliferation, as previously mentioned. Nevertheless, the results of these recent studies (Xu et al., 2004; Waldo et al., 2005a) clearly show that both TBX-1 loss of function and NC ablation lead to a shortened outflow tract; however, the cellular mechanisms underlying this sort of outflow tract underdevelopment need to be further clarified. Furthermore, it must also be pointed out that this sort of modulation exerted by the NC cells on the FGF signaling, as emphasized by Waldo et al. (2005a), seems to represent a conjecture at present and further experimental evidence is needed in support of this hypothesis. Another important aspect of these very recent reports (Waldo et al., 2005b) concerns the differentiation outlined by these authors in terms of aortic sac cellular components. Indeed, these researchers indicate that the SHF provides not only myocardial precursor cells to the developing cardiac outflow tract, but also at later stages smooth muscle precursor cells to the very caudal portion

of the aortic sac, its upper portion being considered by the authors as an entirely NC derivative. It is also important to note that Waldo et al. (2005b) consider the A-P septum to be an entirely NC derivative structure; indeed, they emphasize the fact that the SHF provides primordial smooth muscle cells only to the parietal portions of the caudal aortic sac region, whereas the proximal septal portion of the aortic sac itself (the suprasemilunar valvar aortopulmonary septum), representing postnatally the facing walls of the most proximal intrapericardial portions of the great arteries, is still considered by the authors as an NC derivative. As far as the arterial walls of the semilunar valves are concerned, however, the authors do not provide any specific information about the mechanism by which the caudal septal and parietal portions of the aortic sac contribute to the formation of the arterial surface of the sinuses of Valsalva, as we have questioned and speculated in our discussion concerning truncal septation and arterial valves development. Furthermore, most recent reports by Ward et al. (2005) and Ward and Kirby (2006) emphasize the hypothesis that a short outflow tract, as obtained experimentally by them through SHF ablation and as previously obtained through experimental NC ablation (Yelbuz et al., 2002; Waldo et al., 2005a) and with consequent low SHF cellular output to the conotruncal region, does not allow a normal conotruncal rotation. Based on this hypothesis, the authors explain NC-related conotruncal heart defects such as PTA, tetralogy of Fallot (TF), pulmonary atresia with ventricular septal defect (VSD), and double-outlet right ventricle (DORV) as a consequence of the primary short conotruncal morphology. However, considering these defects in human pathology, we think that a distinction should be made between PTA, TF, and DORV. Indeed, while in PTA the outflow tract is short in the majority of cases, in the aortic dextroposition group (TF) and in DORV conditions, this is not the case. More specifically, as far as TF is concerned, if we exclude the cases with hypoplastic or absent infundibular septum, in the majority of cases the subpulmonary infundibulum is not short but, rather, normally developed along its longitudinal axis, as we see it. More than that, in DORV, excluding the rare cases showing bilateral conal deficiency or absence, in most of the cases a well-developed bilateral infundibular musculature is present and, indeed, the distal right ventricular outflow tract is normally expanded longitudinally. Thus, even if we have to consider that there is a high incidence of conotruncal defects with short infundibular anatomy (hypolplastic or absent infundibular septum) among the neural crestrelated 22q11 deletion syndrome cases (Momma et al., 1995; Marino et al., 1996), in general terms, postnatal morphologic data in human pathology do not seem to show, we think, high incidence of short outflow tract among the conotruncal defects. On the other hand, it is a fact that any time the right ventricle shows a short outflow tract, a total or partial lack of conotruncal rotation and remodeling is inevitably present. Thus, in an attempt of correlative analysis of morphologic data between experimental animal models and human malformations, we think that a poor longitudinal growth of the conotruncal musculature never allows a normal counterclockwise conotruncal rotation, looking from the ventricular side, as the investigations of Ward et al.

CARDIAC OUTFLOW TRACT

(2005) and Ward and Kirby (2006) demonstrate and emphasize. On the other hand, it is also true that this sort of deficient or absent conotruncal rotation occurs in normal conotruncal growth situations as well, such as in the majority of TF and DORV cases. Thus, these data emphasize the fact, we think, that conotruncal hypoplasia is not the only element to cause rotational maldeveopment. Moreover, we found the investigation of Ward et al. (2005) and Ward and Kirby (2006) particularly stimulating and we remain interested to obtain further information about the spiral migratory pathway of the myocardial precursor cells from the right sided portion of the SHF to the back of the left-sided pulmonary portion of the conotruncus within the inner conoventricular flange, as outlined by these authors. Lastly, another recent investigation, by Bajolle et al. (2006), on mouse embryos focuses on and confirms the essential role of the counterclockwise conotruncal rotation, looking from the ventricular side, for normal ventriculoarterial connections to be established. Indeed, previous studies on human embryos (Goor et al., 1972; Lomonico et al., 1986) and chick embryos (Dor and Corone, 1985; Thompson et al., 1987) suggested that the conotruncus undergoes this sort of rotation during its remodeling. The report of Bajolle et al. (2006) shows unequivocally that such a rotation occurs and is produced by movements of the OFT myocardium. More specifically, elongation of the OFT is shown to occur before its rotation, although the authors do not seem to emphasize the elongation as a prerequisite for the rotation itself, as hypothesized by Ward et al. (2005). Furthermore, this study emphasizes the fact that, in the light of the experimental results, the rotation of the conotruncal myocardial wall seems to be linked to the influx of neural crest. In the experimental situations reported by these investigators, in fact, Splotch (Pax3) mutants resulted in PTA and DORV defects, which share lack of conotruncal rotation, and it is well established that normally the Pax3 gene is involved to allow normal NC cell migration from the neural tube down to the pharyngeal apparatus and to the arterial pole of the cardiac tube. It is important to point out that NC cell disruption is well known to interfere also with the septation of the aortic sac and arches and with the septation of the conotruncus, and the PTA malformation is the classical example that recapitulates such a disruption (Kirby et al., 1983). Thus, NC cells seem to exert an important influence on the septation of the conotruncus and, presumably, on its rotation as well, as these latest studies seem to show. In this regard, however, it is noteworthy that while according to Ward et al. (2005), the neural crest influence on this rotatory remodeling of the conotruncus is specifically put in relation to the modulation exerted by the NC cells on the SHF, which ultimately allows proper elongation of the OFT, on the other hand, the neural crest contribution to the conotruncal rotation is not specifically emphasized by Bajolle et al. (2006) in terms of NC-to-SHF modulation. Nevertheless, the role of the NC cells on the conotruncal septation and rotation appears not to be absolute, in the sense that additional factors seem to play an equally important role in this regard. Indeed, a study by Liu et al. (2002) reports a 7% incidence of PTA malformation, where NC cell migration was found to occur normally, but with a Pitx2c mutant situation, and Pitx2c is well established to be a gene involved in left-right laterality signaling. Fur-

941

thermore, the study of Bajolle et al (2006) reports PTA, DORV, and transposition of great arteries (TGA), all situations where lack of conotruncal rotation occurs, in mouse experimental models that showed Pitx2c mutation and normal NC cells supply to the heart. In this respect, then, it is interesting what Bajolle et al. (2006) emphasize in their study regarding the possibility of embryonic left-right laterality signaling disorder in contributing considerably to the production of conotruncal rotational maldevelopment, as the Pitx2c mutation seems to suggest. Furthermore, it is noteworthy to make a consideration regarding the experimental morphologic data obtained by the investigation of Bajolle et al. (2006). Indeed, it is interesting to note that PTA, DORV, and TGA were found in Pitx2c mutation mouse models with a normal NC cell migration and supply to the developing heart. All three malformative conditions share a common lack of conotruncal rotation, although in PTA the conal musculature is almost always entirely absent, whereas in DORV and TGA it is fully developed. However, it is important to point out that, as it occurs in most of the cases in human pathology, while in PTA there is a lack of conotruncal and aortopulmonary septations, in DORV and TGA, in contrast, the conotruncal and aortopulmonary septa are fully formed. On the other hand, it is also important to note that while in TGA and in the majority of PTA cases the left-sided pulmonary portion of the conotruncus translocates leftward from the right ventricle to the left ventricle, such a leftward transfer does not occur in DORV. Indeed, according to Goor and Edwards (1973), DORV and TGA share a lack of conotruncal rotation, with the difference that in TGA the leftward shift of the conoventricular flange transfers the pulmonary conotruncus above the left ventricle, whereas such a shift does not occur in DORV according to this hypothesis. We think that the same mechanism would apply, virtually, to the typical biventricular PTA, where lack of conotruncal rotation, combined with the leftward shift of the left-sided pulmonary portion of the conotruncus would create, potentially, a TGA sort of situation, with the difference that in PTA aortopulmonary and conotruncal septations do not occur usually, whereas in TGA they do. Similarly, considering the unusual forms of PTA arising entirely from the right ventricle, lack of both conotruncal rotation and leftward shift of the left-sided pulmonary conotruncal region would put this defect in a virtual DORV sort of situation. We think that the role of the leftward shift of the conoventricular flange, as originally hypothesized by Goor and Edwards (1973), remains relevant in this sort of embryogenetic mechanisms; however, the intimate biocellular and biophysical processes that allow the rotated subaortic conus, in normal conditions, or the unrotated subpulmonary conus, in TGA and virtually in PTA, to be transferred from the right ventricle to the left ventricle remain poorly known at present, to our knowledge. Thus, we think that these experimental morphologic data reported by Bajolle et al. (2006), if collectively analyzed, seem to show that, in spite of common morphogenetic features related to rotational maldevelopment, a discrepancy exists among these cardiac defects in terms of conotruncal migration, septation, and growth. In this regard, then, further research is most likely to be expected in terms of molecular regulation of the conotruncal development and in terms of possible multifactorial contribution and interplay in such embryogenetic processes.

942

RESTIVO ET AL.

CONCLUSIONS In this review, we have tried to focus on some aspects that we think are not entirely clear yet as far as the development of the cardiac outflow tract is concerned. In considering truncal septation and the correlative development of the arterial semilunar valves, the derivative source of the arterial walls of the sinuses of Valsalva remains unclear, just as it seems unclear the exact mechanism of myocardial cuff retraction and the concomitant caudal translocation of the aortic sac to the arterial valvar roots. Furthermore, the recent knowledge about the SHF opens new frontiers in the understanding of the OFT development. Indeed, further experimental data might substantiate the evidence that the NC cells exert a modulation on the SHF conotruncal myocardial precursors cells as currently hypothesized in the literature. It is also hoped that more data might be obtained as to what extent the NC cells influence the conotruncal septation. More specifically, further information is hoped and expected on whether conotruncal cushion development is modulated by the contribution of additional factors, since PTA malformation, as an example, has been found with normal neural crest migratory pathway in a limited number of cases. It seems to be equally fascinating in future research to obtain more data and information regarding the amount of influence of the NC cells and the SHF on the rotational remodeling of the OFT and its alignment sequels, as recent studies on animal models wish to emphasize. On the other hand, most recent experimental studies on mice seem to suggest the modulatory action of genes, more generally involved in left-right laterality signaling, on the rotation of the conotruncal musculature; further data and information in this direction are most likely to be expected.

LITERATURE CITED Anderson RH, Webb S, Brown NA, Lamers W, Moorman A. 2003. Development of the heart: III, formation of the ventricular outflow tracts, arterial valves, and intrapericardial trunks. Heart 89:1110–1118. Arguello C, De la Cruz MV, Sanchez C. 1978. Ultrastrucural and experimental evidence of myocardial cell differentiation into connective tissue cells in embryonic chick heart. J Mol Cell Cardiol. 10:307–315. Bajolle F, Zaffran S, Kelly RG, Hadchouel J, Bonnet D, Brown NA, Buchingham M. 2006. Rotation of the myocardial wall of the outflow tract is implicated in the normal positioning of the great arteries. Circ Res 98:421–428. Baker AE, Anderson RH. 1981. Pathology of congenital heart disease. Boston: Butterworth. p 309–318. Carr I, Bharati S, Kusnoor I, Lev M. 1979. Truncus arteriosus communis with intact ventricular septum. Br Heart J 42:97–102. Collett RW, Edwards JE. 1949. Persistent truncus arteriosus: a classification according to anatomic types. Surg Clin North Am 29:1245. De la Cruz MV, Gomez CS, Arteaga MN, Arguello C. 1977. Experimental study of the development of the truncus and the conus in the chick embryo. J Anat 123:661–686. Dor X, Corone P. 1985. Migration and torsions of the conotruncus in the chick embryo heart: observational evidence and conclusions drawn from experimental intervention. Heart Vessels 1:195–211. Farrell MJ, Burch JL, Wallis K, Rowley L, Kumiski D, Stadt H, Godt RE, Creazzo TL, Kirby ML. 2001. FGF-8 in the ventral pharynx alters development of myocardial calcium transients after neural crest ablation. J Clin Invest 107:1509–1517. Goor DA, Dische R, Lillehei CW. 1972. The conotruncus: its normal inversion and conus absorption. Circulation 46:375.

Goor DA, Edwards JE. 1973. The spectrum of transposition of the great arteries: with specific reference to the developmental anatomy of the conus. Circulation 48:406. Keith A. 1909. The Hunterian lectures on malformations of the heart. Lancet 12:359–364. Kelly RG, Brown NA, Buckingham ME. 2001. The arterial pole of the mouse heart forms from Fgf10-expressing cells in pharyngeal mesoderm. Dev Cell 1:435–440. Kirby ML, Gale TF, Stewart DE. 1983. Neural crest cells contribute to normal aortico-pulmonary septation. Science 220:1059– 1061. Liu C, Liu W, Palje J, Lu MF, Brown NA, Martin JF. 2002. Pitx2c patterns anterior myocardium and aortic arch vessels and is reqired for local cell movement into atrioventricular cushions. Development 129:5081–5091. Lomonico MP, Moore GW, Hutchins GM. 1986. Rotation of the junction of the outflow tract and great arteries in the embryonic human heart. Anat Rec 216:516–544. Marino B, Digilio MC, Grazioli S, Formigari R, Mingarelli R, Giannotti A, Dallapiccola B. 1996. Associated cardiac anomalies in isolated and syndromic patients with tetralogy of Fallot. Am J Cardiol 77:505– 508. Mjaatvedt CH, Nakaoka T, Moreno-Rodriguez R, Norris RA, Kern MJ, Eisenberg CA, Turner D, Markwald RR. 2001. The outflow tract of the heart is recruited from a novel heart forming field. Dev Biol 238:97–109. Momma K, Kondo C, Ando M, Matsuoka R, Takao A. 1995. Tetralogy of Fallot associated with chromosome 22q11 deletion. Am J Cardiol 76:618–621. Patten BM. 1964. Foundation of embryology, 2nd ed. New York: McGraw-Hill. p 525. Pexieder T. 1975. Cell death in the morphogenesis and teratogenesis of the heart. Adv Anat Embryol Cell Biol 51:100. Ranger AM, Grusby MJ, Hodge MR, Gravallese EM, de la Brousse FC, Hoey T, Mickanin C, Baldwin HS, Glimcher LH. 1998. The transcription factor NF-Atc is essential for cardiac valve formation. Nature 392:186–190. Rosenquist GC, Bharati S, McAllister HA, Lev M. 1976. Truncus arteriosus communis: truncal valve anomalies associated with small conal or truncal septal defects. Am J Cardiol 37:410–412. Schilham MW, Oosterwegel MA, Moerer P, Ya J, de Boer PA, van de Wetering M, Verbeek S, Lamers WH, Kruisbeek AM, Cumano A, Clevers H. 1996. Defects in cardiac outflow tract formation and pro-beta lymphocyte expansion in mice lacking Sox-4. Nature 380:711–714. Thompson RP, Fitzharris TP. 1979a. Morphogenesis of the truncus arteriosus of the chick embryo heart: the formation and migration of mesenchymal tissue. Am J Anat 154:545–556. Thompson RP, Fitzharris TP. 1979b. Morphogenesis of the truncus arteriosus of the chick embryo heart: tissue reorganization during septation. Am J Anat 156:251–264. Thompson RP, Abercrombie V, Wong M. 1987. Morphogenesis of the truncus arteriosus of the chick embryo heart: movements of autoradiographic tattoos during septation. Anat Rec 218:434–440. Van Mierop LHS, Patterson DF, Schnarr WR. 1978. Pathogenesis of persistent truncus arteriosus in light of observations made in a dog embryo with the anomaly. Am J Cardiol 41:755–762. Van Mierop LHS, Patterson DF. 1980. Pathogenesis of conotruncal defects in the keeshond dog. In: Van Praagh R, Takao A, editors. Etiology and morphogenesis of congenital heart disease. New York: Futura Publishing. p 177–195. Waldo KL, Kumiski D, Wallis KT, Stadt HA, Hutson MR, Platt DH, Kirby ML. 2001. Conotruncal myocardium arises from a secondary heart field. Development 128:3179–3188. Waldo KL, Hutson MR, Stadt HA, Zdanowicz M, Zdanowicz J, Kirby ML. 2005a. Cardiac neural crest is necessary for normal addition of the myocardium to the arterial pole from the secondary heart field. Dev Biol 281:66–77. Waldo KL, Hutson MR, Ward CC, Zdanowicz M, Stadt HA, Kumiski D, Abu-Issa R, Kirby ML. 2005b. Secondary heart field contributes myocardium and smooth muscle to the arterial pole of the developing heart. Dev Biol 281:78–90.

CARDIAC OUTFLOW TRACT Ward C, Stadt H, Hutson M, Kirby ML. 2005. Ablation of the secondary heart field leads to tetralogy of Fallot and pulmonary atresia. Dev Biol 284:72–83. Ward C, Kirby ML. 2006. The secondary heart field: understanding conotruncal defects from a developmental perspective. Curr Cardiol Rev 2:5–20. Watanabe M, Jafri A, Fisher SA. 2001. Apoptosis is required for the proper formation of the ventriculo-arterial connections. Dev Biol 240:274–288. Xu H, Morishima M, Wylie JN, Schwartz RJ, Bruneau BG, Lindsay EA, Baldini A. 2004. TBX1 has a dual role in the morpho-

943

genesis of the cardiac outflow tract. Development 131:3217– 3227. Ya J, van den Hoff MJ, de Boer PA, Tesink-Taekema S, Franco D, Moorman AF, Lamers WH. 1998a. Normal development of the outflow tract in the rat. Circ Res 82:464–472. Ya J, Schilham MW, de Boer PA, Moorman AF, Clevers H, Lamers WH. 1998b. Sox-4 deficiency syndrome in mice is an animal model for common trunk. Circ Res 83:986–994. Yelbuz TM, Waldo KL, Kumiski DH, Stadt HA, Wolfe RR, Leatherbury L, Kirby ML. 2002. Shortened outflow tract leads to altered cardiac looping after neural crest ablation. Circulation 106:504–510.