The heart in higher vertebrates develops from a simple tube into a complex organ with four chambers specialized for efficient pumping at pressure. During this period, there is a concomitant change in the level of myocardial organization. One important event is the emergence of trabeculations in the luminal layers of the ventricles, a feature which enables the myocardium to increase its mass in the absence of any discrete coronary circulation. In subsequent development, this trabecular layer becomes solidified in its deeper part, thus increasing the compact component of the ventricular myocardium. The remaining layer adjacent to the ventricular lumen retains its trabeculations, with patterns which are both ventricle- and species-specific. During ontogenesis, the compact layer is initially only a few cells thick, but gradually develops a multilayered spiral architecture. A similar process can be charted in the atrial myocardium, where the luminal trabeculations become the pectinate muscles. Their extent then provides the best guide for distinguishing intrinsically the morphologically right from the left atrium. We review the variations of these processes during the development of the human heart and hearts from commonly used laboratory species (chick, mouse, and rat). Comparison with hearts from lower vertebrates is also provided. Despite some variations, such as the final pattern of papillary or pectinate muscles, the hearts observe the same biomechanical rules, and thus share many common points. The functional importance of myocardial organization is demonstrated by lethality of mouse mutants with perturbed myocardial architecture. We conclude that experimental studies uncovering the rules of myocardial assembly are relevant for the full understanding of development of the human heart.

Normal stages of intracardiac development of the mouse heart are illustrated by a collection and analysis of two complementary SEM views, the septal and parietal halves of the right ventricle and conotruncus. These views are particularly suitable for understanding the septation of the outflow tract of the heart. They represent references for the studies of the pathogenesis of conotruncal malformations. The analysis is based on 90 hearts between 11ed and 15ed at intervals of 8 hours with an additional stage at 16ed. They were prepared by perfusion fixation, microdissection, and critical point drying and were examined in SEM. The following main features of the intracardiac morphogenesis were observed: 1) the presence of two spirally positioned conotruncal ridges, their disto-proximal fusion, and the formation of the semilunar valves at their distal part; 2) the trabecular transformation of the conotruncal wall; and 3) the presence of two different interventricular foramina (FIV II and FIV III), with the final closure of FIV III at 14ed16h-15ed. As a result of these observations, our description of the outflow tract septation and its interpretation are different from other conceptions based on mechanisms such as bulbar shift, bulbar absorption, torsions, or transfer of the aorta into the left ventricle.

Normal development of the mouse embryonic heart was studied at the organ level using microdissection and scanning electron microscopy (SEM). Altogether 225 embryos, sampled at 8-hour intervals between 11ed (ed = embryonic day; day of vaginal plug = 1ed) and 15ed were collected. Their hearts were fixed by high flow-low pressure perfusion, microdissected, and observed in SEM. Standardized frontal, right profile, and left profile SEM micrographs were obtained and analyzed. The main purpose of this study was to create a series of normal stages of mouse cardiac development as a reference for ongoing studies in experimental cardiac teratology (e.g., in fetal mouse trisomies). Comparisons with chick, human, and dog embryonic hearts, prepared using the same technique, show that the mouse embryonic heart is characterized by a relatively deep interventricular sulcus. The absence of a conoventricular sulcus in the mouse results in poor definition of the boundary between the conus and the right ventricle. The external separation of the aorta and the pulmonary artery is evident from 13ed onward. The respective positions of the great arteries (aorta dextroposterior, pulmonary artery sinistroanterior) does not change until the end of cardiac organogenesis (15ed in the mouse).

Carbonic anhydrase II localization was studied in mouse embryonic and fetal hearts for better understanding of the functions of this enzyme during cardiac organogenesis. Immunocytolabelling was performed on serial sections of frozen hearts after one night's fixation in 4% paraformaldehyde. In the earliest stages studied, 10, 11 and 12 ed (ed = embryonic day; vaginal plug = day 1), a sharp decrease of labelled cells was observed in the endocardium form which cushion-tissue mesenchyme is derived. During the same period, differences in the decreasing frequencies of labelled cells were also observed between three different cushion-tissue mesenchyme localizations: immunostained cells were abundant in the atrioventricular cushions, less numerous in the proximal part of the conotruncal ridges and rare in their distal part. From 13 ed their repartition was more regular along the conotruncus. From 13 to 16 ed the signal was also present in a peculiar region of the myocardium: the anterior and left walls of the left ventricle. At the 18 and 20 ed labelling was found only in some endothelial cells of coronary vessels, particularly in the interventricular septum. The pattern of expression of carbonic anhydrase II in activated endothelial cells and endothelial-derived mesenchyme cells of the cardiac cushion tissue, strongly suggests that this isoenzyme can be a useful marker for a subpopulation of endothelial cells and cells derived from this endothelium that morphologically express signs of active cell behavior (e.g., invasion, migration, proliferation).

No abstract