Search

Background: Nitric oxide (NO) is a multifunctional gaseous molecule that regulates various physiological functions in both neuronal and non-neuronal cells. NO is synthesized by nitric oxide synthases (NOSs), of which three isoforms have been identified. Neuronal NOS (nNOS) and endothelial NOS (eNOS) constitutively produce low levels of NO as a cell-signaling molecule in response to an increase in intracellular calcium concentration. Recent data have revealed a predominant role of eNOS in both angiogenesis and vasculogenesis., Methods: The immunohistochemical localization of nNOS and eNOS was investigated during embryonic and fetal ocular vascular development from 7 to 21 weeks gestation (WG) on sections of cryopreserved tissue., Results: eNOS was confined to endothelial cells of developing vessels at all ages studied. nNOS was prominent in nuclei of vascular endothelial and smooth muscle cells in the fetal vasculature of vitreous and choriocapillaris. nNOS was also prominent in the nuclei of CXCR4(+) progenitors in the inner retina and inner neuroblastic layer., Conclusions: These findings demonstrate co-expression of n- and eNOS isoforms in different compartments of vasoformative cells during development. Nuclear nNOS was present in vascular and nonvascular progenitors as well as endothelial cells and pericytes. This suggests that nNOS may play a role in the transcription regulatory systems in endothelial cells and pericytes during ocular hemo-vasculogenesis, vasculogenesis, and angiogenesis.

Hepatic blood vessels consist of the hepatic artery and three types of venous channels (the portal veins, the sinusoids, and the hepatic veins). This study was undertaken to analyze, by immunohistochemistry, connexin expression throughout the vascular development of the fetal mouse liver with special attention being given to portal vein development. In the adult liver, connexin37 and connexin40 were expressed in the endothelium of the portal vein and hepatic artery, but not in those of the hepatic vein and sinusoids. Connexin43 was expressed in mesothelial cells and smooth muscle cells of the portal veins. The preferential expression of connexin37 and connexin40 in portal veins was seen throughout liver development, including its primordium formation stage (10.5-day or 11.5-day stage), although connexin37 expression was transiently seen in free nonparenchymal cells in fetal stages. The differentiation of each blood vessel in the hepatic vascular system may occur in early developmental stages, soon after hepatic primordium formation.

Our understanding of the regulation of vascular development has exploded over the past decade. Prior to this time, our knowledge of vascular development was primarily based on classic descriptive studies. The identification of stem cells, lineage markers, specific growth factors and their receptors, and signalling pathways has facilitated a rapid expansion in information regarding details of the mechanisms that govern development of the vascular system.

The vertebrate vascular system is essential for the delivery and exchange of gases, hormones, metabolic wastes and immunity factors. These essential functions are carried out in large part by two types of anatomically distinct blood vessels, namely arteries and veins. Previously, circulatory dynamics were thought to play a major role in establishing this dichotomy, but recently it has become clear that arterial and venous endothelial cells are molecularly distinct even before the output of the first embryonic heartbeat, thus revealing the existence of genetic programs coordinating arterial-venous differentiation. Here we review some of the molecular mechanisms involved in this process.

In contrast to the extensive molecular and functional characterization of blood vascular endothelium, little is known about the mechanisms that control the formation and lineage-specific differentiation and function of lymphatic vessels. The homeobox gene Prox1, the vertebrate homologue of the Drosophila prospero gene, has been recently identified to be required for the induction of lymphatic vascular development from preexisting embryonic veins, and studies in Prox1-deficient mice have confirmed Florence Sabin's original hypothesis about the origin of the lymphatic vascular system from embryonic veins. The recent establishment of cell culture models for the selective propagation of blood vascular and lymphatic endothelial cells, together with the findings that these cells maintain their lineage-specific differentiation in vitro, has led to the discovery that Prox1 expression is sufficient to induce a lymphatic phenotype in blood vascular endothelium. Ectopic expression of Prox1 downregulated blood vascular-associated genes and also upregulated some of the known lymphatic endothelial cell markers. Together, these studies suggest that the blood vascular phenotype represents the default endothelial differentiation and they identify an essential role of Prox1 in the program specifying lymphatic endothelial cell fate.

The hypothesis of astroglial cell involvement in prenatal setting up of the blood-brain barrier (BBB) has been examined by producing glial degeneration in cerebellum of chicken embryos submitted to the action of gliotoxin 6-aminonicotinamide (6-AN), which was applied onto the embryo chorioallantoic membrane during both early and late embryonic development. The effects of 6-AN on the cerebellum astroglial cells and microvessels were analysed under the light microscope by immunostaining for 3CB2 (chick-specific glial marker) and HT7 (chick-specific marker of BBB-provided brain endothelia), under the electron microscope, as well as by the vascular permeability tracer horseradish peroxidase. The results, showing good suitability of the 6-AN model also when applied in early embryonic development, demonstrated a correlation between perivascular glia depletion and endothelial barrier impairment and suggested that astroglia play a role in the BBB prenatal differentiation.

The cardiovascular system is the first functional organ in the vertebrate embryo, and many organs start to develop adjacent to cells of the cardiovascular system. Endothelial cells (EC) form the inner cell lining of blood vessels and represent the major cell type that interacts with developing organs. On the one hand, EC provide organs with signals. These signals determine the location, differentiation and morphology of an organ. On the other hand, EC receive signals from the organ-specific cell types. Such signals give EC organ-specific features that the organ needs to interact with the circulatory system. This review provides the reader with specific examples of an interdependent development of organs and blood vessels.

The lymphatic endothelium has mostly been thought to be derived by sprouting from specialized veins. Recently it has been shown that mice deficient for the homeobox transcription factor Prox1 are practically devoid of lymphatics. We have studied the expression of Prox1 mRNA and protein in chick embryos and human fetuses. In the chick, Prox1 is expressed in specific compartments of all germ layers. In the ectoderm, it is found in the neural tube, trigeminal, spinal and sympathetic ganglia and the retina, and also in placodal structures such as the lens, olfactory, otic, facial, glossopharyngeal and vagal placodes, and the apical ectodermal ridge. In the endoderm, Prox1 is a marker of hepatocytes, bile duct and pancreatic epithelium. In the mesoderm, weak expression is observed in cardiomyocytes, and strong expression in lymphatic endothelium. Identical expression domains are found in 19-week-old human fetuses. In day 6.5 chick embryos, there are several sites of contact of lymphatics with the jugular vein, which has a mixed endothelium of Prox1-positive and -negative cells. The only non-lymphatic endothelial cells expressing Prox1 are found on the concave side of the cardiac valves. To further analyse development of lymphatics, we studied early chick embryos and observed scattered Prox1-positive cells in the dermatome, giving rise to Prox1-positive lymphatic networks during subsequent development. Furthermore, the anlagen of the posterior lymph sacs and the paired thoracic duct can already be observed in day-4 chick embryos. Our studies show that lymphatics develop much earlier than previously described, and they mostly do not seem to be derived by sprouting from veins. In contrast, lymphangioblasts are present in the deep and superficial compartments of the early mesoderm, independently giving rise to the deep and superficial lymphatics.

In its function of separating pulmonary and systemic arterial blood flow, the ductus arteriosus, which connects both circuits, either closes permanently at a certain stage in development or attains a capacity to close and reopen depending on the physiological needs in certain species. In air-breathing vertebrates varying from lungfish to mammals, the ductus arteriosus derives from the sixth pharyngeal arch artery, and in preparation for its specific task, undergoes its own unique differentiation programme, starting early in development. To date, the mechanisms involved in defining this unique status, as compared to the other great arteries, are unclear. This review clarifies some of the elusiveness of the ductus arteriosus. It includes a comparative description of this artery in species exemplifying the different classes of air-breathing vertebrates, and illustrates similarities and differences in morphogenesis and closure mechanisms among the species. It also deals with possible influences of vascular innervation and with congenital anomalies in which the ductus arteriosus is involved. New data suggest that HOXB5 expression in the neural crest along the dorsal half of the sixth arch artery may be involved in the instigation of ductus arteriosus differentiation.

The aim of this work was to address spatiotemporal and morphologic patterns of coronary artery development in rats, based on immunohistochemical and ultrastructural studies of hearts at different stages of prenatal development. Griffonia simplicifolia I lectin and alpha-smooth muscle antibody were used to demonstrate endothelial cells and/or their precursors and smooth muscle cells, respectively. Ultrastructural examination was performed on ED14-16 hearts to study the morphology of the developing coronary arteries in different regions of the truncus arteriosus and adjacent myocardium. On ED14 endothelial-like cells present within the mesenchyme surrounding the outflow tract penetrated the aortic wall and the truncoconal proximal myocardium. On ED15 these penetrating cells formed vascular clusters, which were the first signs of presumptive vascular channels. Development of the coronary artery proceeded by coalescence of discontinous vascular clusters, formation of the lumen (vascular channels) and establishing a connection of the proximal part with the aorta. The second layer of cells around vascular channels (embryonic media) consisted of mesenchymal cells that were attracted to the immature vessel and were first seen on ED15. At this time no lumenized connection of the coronary artery with the aorta has been seen. After the lumenized connection of the coronary artery with the aorta had been established perivascular cells of the media started to differentiate into vascular smooth muscle, as was shown by alpha-smooth muscle actin-staining. Further development and differentiation of the media and adventitia proceeded distally (towards the apex).

The presence of non-contractile smooth muscle cells within the arterial wall raises questions as to their origin and function. These cells abound within the aortae of murine and porcine neonates, but are also present within the intimal and medial layers of adult arteries. They are largely devoid of smooth muscle-associated proteins and manifest an epithelioid form. Their morphological resemblance to endothelial cells prompted us to explore this potential relationship and to investigate their angiogenic properties in three-dimensional collagen gels. Using well-characterized smooth muscle cell lines, displaying either the intima-like (epithelioid) or media-like (spindle-shaped) morphology, we were able to show that intima-like cells share several features in common with endothelial ones and can transform into a media-like phenotype, whereby they irreversibly lose their characteristic pattern of protein expression. Intima-like, but not media-like, vascular smooth muscle cells are capable of forming capillary tubes, and, in co-cultures, can induce media-like ones to participate in this process. Such capillaries consist of a randomly-organized, mixed population of endothelial cells with intima-like or media-like smooth muscle ones. The functional significance of this diversity in smooth muscle cell type is not well understood, but phenotypic plasticity could conceivably figure as an important adaptive response to changes in the local environment.

Endoglin is a component of the receptor complex for transforming growth factor (TGF)-beta1 and TGF-beta3. We analysed its expression by immunohistochemistry in human embryos at 4-8 weeks of gestation and in hearts ranging from 4-13 weeks old. We compared endoglin distribution with that of TGF-beta receptors type I (TbetaR-I), type II (TbetaR-II) and betaglycan. Endoglin was found on endothelial cells in all tissues examined, consistent with its expression in adult blood vessels. TbetaR-I, TbetaR-II and betaglycan were observed on most cell types and had an overall similar pattern of distribution. Endoglin was detected on the endocardium as early as 4 weeks, but was absent from myocardium. It was present at high levels on the endocardial cushion tissue mesenchyme from 5-8 weeks' gestation, during heart septation and valve formation, and subsequently decreased as the valves matured. Endoglin expression in heart extracts was confirmed by Western blot analysis. TbetaR-I, TbetaR-II and betaglycan were mostly found on cardiac myocytes, but were detectable at low levels on endocardium. They were expressed transiently on cushion mesenchyme, albeit at much lower levels than endoglin. All four components of the TGF-beta receptor complex were detected by RT-PCR in embryonic heart. Thus transient up-regulation of the components of the TGF-beta receptor complex, and particulartly of endoglin, is associated with heart septation and valve formation during early human development.

We have studied the expression of Quek1 and Quek2 (VEGFR-2 and VEGFR-3, respectively) in quail embryos from day 2 to day 16 by in situ hybridization with digoxigenin-labelled riboprobes on whole-mounts and paraffin sections. Parallel sections were also stained with the QH1 antibody to detect all endothelial cells and with an antibody against alpha-smooth-muscle-actin to reveal the media of blood vessels. Quek1/VEGFR-2 is a marker of blood-vascular and lymphatic endothelial cells throughout development. In 2-day-old embryos, it is expressed in the intra-embryonic vascular plexus, in cells (most probably angioblasts) located in the paraxial head mesoderm and in the somites, and caudo-laterally from Hensen's node. Thereafter, until about day 9, Quek1 is expressed in all endothelial cells. Cells positive and negative for Quek1 can later be found within the same vessel. Quek1 is additionally expressed in lymphatic endothelial cells. Occasionally, some non-endothelial cell types express Quek1. Quek2/VEGFR-3 is also a marker of endothelial cells; however, its expression pattern differs from that of Quek1. In 2-day-old embryos, Quek2 is expressed in the notochord and the intra-embryonic vascular plexus. Whereas all endothelial cells are Quek2-positive in 3-day-old embryos, expression is subsequently reduced to a subset of endothelial cells: arteries become Quek2-negative and then expression of Quek2 is limited to a few vessels that appear to be lymphatic. Endothelial cells of lymph nodes and the periaortal lymphatic vessels are Quek2-positive in later stages. A few non-endothelial cells express Quek2.

In the midline of the embryo an invisible barrier exists that keeps endothelial cells from migrating to the contralateral side. Interspecific grafting experiments between chick and quail were carried out in order to investigate the role of the axial structures in maintaining this barrier. The quail endothelial cells of the graft were therefore stained with QH1 antibody. In all experimental series quail paraxial mesoderm was used as a source of endothelial cells. First, a quail somite was transplanted either ipsilaterally or contralaterally. The results not only show the existence of laterality in the distribution pattern, but also demonstrate that the laterality does not depend on the origin of the graft but on the environment of the host embryo. Laterality in the distribution pattern of endothelial cells means that the endothelial cells of the two body halves migrate independently and do not change from one side to the other. Single cells do not know whether they are cells from the right or from the left half of the body. In the next series of experiments axial structures were removed in order to modify the barrier. In addition, paraxial mesoderm was exchanged with the corresponding quail tissue in order to determine the migration behaviour of the grafted endothelial cells. The removal of the neural tube does not influence the barrier. After notochordectomy, however, the endothelial cells exhibited a balanced distribution pattern over both halves of the embryo. We concluded that the notochord forms a barrier for endothelial cells that presumably operates on the basis of chemical substances. It is conceivable that our results can explain the lateralization of illnesses of the vascular system, as the Klippel-Trénaunay syndrome or the Sturge-Weber syndrome.

In a previous study of the hemangioblastic capacity of lateral plate mesoderm, we showed that the endoderm-associated splanchnopleural layer is capable of giving rise to both endothelial and hemopoietic cells while the ectoderm-associated somatopleural layer is not (Pardanaud and Dieterlen-Lièvre 1993a). In order to complete the inventory of territories able to produce these two cell lineages, we assayed the paraxial mesoderm, and report the results here. Quail somites or segmental plates were treated with mab QH1+complement in order to eliminate attached aortic endothelial cells, which cling to the ventral aspects of these structures. They were grafted in the limb bud or the coelom of chick host, since these sites promote the differentiation of endothelial and hemopoietic cells, respectively. Vascular development and hemopoietic cell emergence were analyzed using QH1 immunocytology. Segmental plate and somites both produced abundant endothelial cells. In addition, the segmental plate gave rise to small groups of hemopoietic cells when grafted in the coelom.

The development of the vertebrate limb requires the formation of a normal vasculature to nurture the soft and hard tissue phenotypes. The pattern of embryonic limb bud vessels has been extensively studied, but little is known about the permeability characteristics of the developing circulation. In the present study, the microvascular endothelial cell phenotype was examined by in vivo confocal microscopy following the systemic injection of a graded series of fluorescent dextrans (40,000, 70,000, 150,000 molecular weight) into chick embryos at stages 21-23 in order to determine how selective is the endothelial lining of microvessels as a partition between the blood vessels and the interstitium. Videodensitometry, over a gray scale range of 0-255, was used to quantitate the amount of tracer found within the interstitial compartment of the limb. The tracers of larger molecular weight (70,000, 150,000) were confined exclusively to the vascular lumina, whereas that of smaller molecular weight (40,000) was found to cause perivascular brightening due to extravasation into the surrounding interstitium. The reported differences in permeability were not dependent upon the stage of the embryo used in this study, but were due to the size of the tracer. These data indicate that embryonic wing microvessels demonstrate permselectivity to macromolecular efflux across the endothelium. The present results provide a basis for additional studies concerned with the dynamic characteristics of the limb microvasculature and challenge our concepts about the role of diffusible morphogens in vertebrate limb development.

The effect of vascular endothelial growth factor (VEGF165) on the chorioallantoic membrane (CAM) of 13-day-old chick embryos was studied. The factor was applied in doses of 0.5-4 micrograms for a period of up to 4 days. Macroscopical, histological and immunohistological studies were carried out. The localization of the factor was examined with an anti-VEGF antibody. The mitogenicity of VEGF165 and basic fibroblast growth factor (bFGF) were studied by means of the BrdU-anti-BrdU method. Furthermore, the effect of heparin alone and in combination with VEGF165 was investigated. VEGF165 specifically induces angiogenesis in doses of 0.5 microgram and more. A brush-like formation of blood vessels can be seen in the region of the precapillary vessels. Angiogenesis also takes place in the region of the capillaries and the venules. Histologically we found indications of sprouting as well as of intussusceptive capillary growth. The presence of the factor in the application area could be demonstrated with the anti-VEGF antibody for a period of 3 days. The factor is located in the chorionic epithelium and the intraepithelial capillaries. The BrdU-studies show that VEGF165 induces strong endothelial cell proliferation, whereas bFGF elicits fibrocyte proliferation and minor endothelial cell proliferation. Heparin induces squamous metaplasia of the chorionic and allantoic epithelium in combination with an aggregation of fibrocytes. We could not detect any enhancement of VEGF165 by heparin.

The role played by perivascular astrocytes in neural vessel maturation was investigated in microvessels of the chick embryo optic tectum. Three-dimensional reconstructions and quantitative analyses were made, and permeability was studied. On embryonic days 14-16, 12.5% of the microvessel wall is surrounded by astrocyte endfeet which, in most cases (82%), are located under endothelium junctions; the latter, at this stage, partly prevent the extravascular escape of the marker horseradish peroxidase. On days 18-21, the astrocyte processes form a nearly complete perivascular sheath enveloping 96% of the microvessel perimeter; the junctions of the endothelial cells are much wider and impermeable owing to extensive fusion of the endothelial plasma membranes. This investigation suggests a close relationship between the perivascular arrangement of glia and differentiation of the endothelium tight junctions and indicates that the morphofunctional maturation of the latter takes place progressively during the prenatal organogenesis of the chick central nervous system.

During organogenesis, endothelial cells develop through two different mechanisms: differentiation of intrinsic precursors in organ rudiments constituted of mesoderm associated with endoderm, and colonization by extrinsic precursors in organs constituted of mesoderm associated with ectoderm (Pardanaud et al. 1989). On the other hand, both types of rudiment are colonized by extrinsic hemopoietic stem cells. In the present work we extend our former study by investigating the hemangioblastic (i.e. hemopoietic and angioblastic) potentialities of primordial germ layers in the area pellucida during the morphogenetic period. By means of interspecific grafts between quail and chick embryos, we show that splanchnopleural mesoderm gives rise to abundant endothelial cells, and to numerous hemopoietic cells in a permissive microenvironment, while somatopleural mesoderm produces very few cells belonging to these lineages, or none. Thus we confirm that the angioblastic capacities of the mesoderm differ radically, depending on its association with ectoderm or endoderm. Furthermore, at this embryonic period, both endothelial and hemopoietic potentialities are displayed by splanchnopleural mesoderm. However the site of emergence of intraembryonic hemopoietic stem cells appears spatially restricted by comparison to more widespread angioblastic capacities.

Cytokinetic studies on the aortic endothelium using the BrdU/anti-BrdU-method were carried out on 2.5- to 6-day chick and quail embryos. The mitotic activity of the aortic endothelium is related temporally to the age of the avian embryo and spatially to the embryonic region where the aorta originates. The mitotic activity of the aortic endothelium decreases with increasing age of the embryos. In the limb buds, however, the mitotic rate of the aortic endothelial cells increases independently of the age of the embryo. This increase in the mitotic activity of the aortic endothelium at the appropriate levels coincides with the vascularization of the outgrowing limb buds. We concluded therefore that the aortic endothelium probably supplies endothelial cells for the formation of limb vessels at this stage. Thus our results suggest that angiogenesis (sprouting of capillaries from pre-existing vessels) takes place during limb vascularization in avian embryos. On the other hand, immunohistochemical studies with QH-1 or MB-1 antibody show, beside a capillary network in the central core of the wing bud, individual immunolabelled cells of mesenchymal character within the primarily avascular subectodermal region from the onset of vascularization onwards. We suggest that these cells have partly to be regarded as endothelial precursor cells, which have differentiated in situ from the local limb mesenchyme, and which will contribute to the developing vascular plexus. This means that not only angiogenesis, but also vasculogenesis (in situ from mesenchymal precursors differentiated endothelial cells) appears to be involved in limb vessel formation.

The microvascular endothelium of the chick wing bud at stages 22, 27, and 32 was evaluated by ultrastructural morphometry. The rationale for this study is based on the hypothesis that endothelial cells exhibit variation in structure and function during cytodifferentiation. The microvessels had a luminal diameter range such that they were classified as capillaries. The thin continuous endothelium was devoid of a basal lamina. The endothelium had a very small number of plasmalemmal vesicles; vacuoles were however present for all stages and in some cases were abundant. The temporal findings were that endothelial cell thickness increases, plasmalemmal vesicle densities decrease, and the densities of cytoplasmic vacuoles increase. The spatial results were that endothelial cells in proximal regions of the limb have a greater thickness, contain fewer vesicles and have more vacuoles than those in distal regions. In general, these results indicate that endothelial ultrastructural heterogeneity occurs within a 3 1/2 day time-span of wing bud development. The discussion considers the results with regard to recent reports on endothelial cell heterogeneity.

According to previous studies, a process of endothelial activation seems to be occurring in the chick embryo between days 7 and 18. Also, endothelial cells respond to collagen as a substratum between 12 and 18 days, and this response diminishes until it almost disappears after birth. In the present study, aortas from chick embryos (days 7 to 21), and from chicks (14 days posthatching) were used. The results obtained by the freeze-fracturing technique, showed that between days 12 and 14 the intramembranous particles were aggregated into linear or clustered arrays in the fracture P-face of endothelial cells. This could signify that some kind of gap junction-like coupling may occur between adjacent endothelial cells. Our results also indicate that in advanced stages (21-day-old chick embryos and 14-day-old chicks) the growth of small aggregates into larger aggregates or plaques could occur. In addition to gap junctions, the presence of macular and linear tight junctions, reported as focal tight junctions (day 14 of development) macular and linear tight junctions with free-ending strands oriented parallel to one another (21 days) and smooth contoured ridges (14 days post-hatching) were observed. This sequence of changes may represent a development from linear to macular, to a more occluding arrangement, and may also reflect an endothelial cell polarization. Histochemical study of proteoglycans was done by using cuprolinic blue according to the critical electrolyte concentration method. Cuprolinic blue-positive granular, elongated and microfibrillar materials were found in the subendothelial region, forming a meshwork that occupies the extracellular space. Qualitative and quantitative changes were observed both in proteoglycans and in other extracellular matrix components throughout development, suggesting an increase in extracellular matrix complexity. These results lead us to suggest that the assembly of a more complex extracellular matrix, concomitantly with the formation of intercellular junctions during development, might influence the polarization of endothelium in the aorta of the chick embryo.

In the present paper we used a method whereby some of the cellular events that take place in the aortic wall during chick embryo development can be studied in vitro. Collagen gels were utilized to culture endothelial cells obtained through nonenzymatic means from aortic explants isolated from 12- to 14-day-old chick embryos. These cells were characterized by morphological and immunocytochemical criteria. After 72 h, explanted endothelial cells from embryonic aorta formed a monolayer of polygonal cells, which gave rise to elongated cells that migrate into the collagen gel. These cells are similar to those of mesenchyme-like cells observed in vivo at the subendothelial region of 14-day-old chick embryonic aorta. In long-term cultures, some of these cells acquired features either of synthetic smooth muscle cell phenotype, or of fibroblast-like cells very similar to those found in developing aorta. These results indicate that the culture of explants of chick embryo aorta on three-dimensional collagen gel is a valuable system for studying some of the complex morphogenetic events that occur during the development of the aorta.