Cryptic responses to tissue manipulations in avian embryos.

Experimental embryology performed on avian embryos combines tissue manipulations and cell-labeling methods with increasing opportunities and demands for critical assays of the results. These approaches continue to reveal unexpected complexities in the normal patterns of cell movement and tissue origins, documentation of which is critical to unraveling the intricacies of cell and tissue interactions during embryogenesis. Viktor Hamburger's many pioneering contributions helped launch and promote the philosophical as well as technical elements of avian experimental embryology. Furthermore, his scholarship and profoundly positive presence influenced not just those of us fortunate to have trained with him, but several generations of developmental biologists. The first part of this article presents examples of the opportunities and rewards that have occurred due to his influences. Surgical manipulation of avian embryonic tissues always introduces a greater number of variables than the experimenter can control for or, often, readily identify. We present the results of dorsal and ventral lesions of hindbrain segments, which include defects in structures within, beside, and also at a considerable distance from the site of lesion. Extramedullary loops of longitudinal tract axons exit and re-enter the neural tube, and intra-medullary proliferation of blood vessels is expanded. Peripherally, the coalescence of neural crest- and placode-derived neuroblasts is disrupted. As expected, motor neurons and their projections close to the sites of lesion are compromised. However, an unexpected finding is that the normal projections of cranial nerves located distant to the lesion site were also disrupted. Following brainstem lesions in the region of rhombomeres 3, 4 or 5, trigeminal or oculomotor axons penetrated the lateral rectus muscle. Surprisingly, the ability of VIth nerve axons to reach the lateral rectus muscle was not destroyed in most cases, even though the terrain through which they needed to pass was disrupted. These axons typically followed a more ventral course than normal, and usually, the axons emerging from individual roots failed to fasciculate into a common VIth nerve, which suggests that each rootlet contains pathfinder-competent axons. The lesson from these lesions is that surgical intervention in avian embryos may have substantial effects upon tissues within, adjacent to, and distant to those that are being manipulated.

LHRH neuronal migration: heterotypic transplantation analysis of guidance cues.

During embryonic development, the olfactory placode (OP) differentiates into the olfactory epithelium (OE). Luteinizing hormone-releasing hormone (LHRH) neurons migrate out of the OE in close association with the olfactory nerve (ON) to the telencephalon. LHRH neuronal migration and ON extension to the telencephalon may be independent events which are correlated but do not represent a causal relationship. However, we hypothesize that LHRH neurons are dependent on ON axons to migrate to the brain. To test this hypothesis, we ablated the right trigeminal placode and replaced it with an OP from another chick embryo. After several days' additional incubation, the embryos were fixed, sectioned, and immunostained with antibodies against LHRH or N-CAM. The ectopic OPs were well integrated into the host and developed into relatively normal appearing OEs. The ONs extended from the OE to several different sites: the lateral rectus of the eye, the ciliary ganglion, and the trigeminal ganglion. In all cases, LHRH neurons were found in the OE and ON, regardless of where the ON terminated. When the ON extended to the trigeminal ganglion, LHRH neurons could clearly be seen entering the metencephalon. Our results support the idea that LHRH neurons are dependent on the ON for guidance as they appear to follow the nerve even when it extends away from the brain. The cues which direct the ON and LHRH neurons to the telencephalon do not appear to be unique to this brain region.

Differentiation of avian craniofacial muscles: I. Patterns of early regulatory gene expression and myosin heavy chain synthesis.

Myogenic populations of the avian head arise within both epithelial (somitic) and mesenchymal (unsegmented) mesodermal populations. The former, which gives rise to neck, tongue, laryngeal, and diaphragmatic muscles, show many similarities to trunk axial, body wall, and appendicular muscles. However, muscle progenitors originating within unsegmented head mesoderm exhibit several distinct features, including multiple ancestries, the absence of several somite lineage-determining regulatory gene products, diverse locations relative to neuraxial and pharyngeal tissues, and a prolonged and necessary interaction with neural crest cells. The object of this study has been to characterize the spatial and temporal patterns of early muscle regulatory gene expression and subsequent myosin heavy chain isoform appearance in avian mesenchyme-derived extraocular and branchial muscles, and compare these with expression patterns in myotome-derived neck and tongue muscles. Myf5 and myoD transcripts are detected in the dorsomedial (epaxial) region of the occipital somites before stage 12, but are not evident in the ventrolateral domain until stage 14. Within unsegmented head mesoderm, myf5 expression begins at stage 13.5 in the second branchial arch, followed within a few hours in the lateral rectus and first branchial arch myoblasts, then other eye and branchial arch muscles. Expression of myoD is detected initially in the first branchial arch beginning at stage 14.5, followed quickly by its appearance in other arches and eye muscles. Multiple foci of myoblasts expressing these transcripts are evident during the early stages of myogenesis in the first and third branchial arches and the lateral rectus-pyramidalis/quadratus complex, suggesting an early patterned segregation of muscle precursors within head mesoderm. Myf5-positive myoblasts forming the hypoglossal cord emerge from the lateral borders of somites 4 and 5 by stage 15 and move ventrally as a cohort. Myosin heavy chain (MyHC) is first immunologically detectable in several eye and branchial arch myofibers between stages 21 and 22, although many tongue and laryngeal muscles do not initiate myosin production until stage 24 or later. Detectable synthesis of the MyHC-S3 isoform, which characterizes myofibers as having "slow" contraction properties, occurs within 1-2 stages of the onset of MyHC synthesis in most head muscles, with tongue and laryngeal muscles being substantially delayed. Such a prolonged, 2- to 3-day period of regulatory gene expression preceding the onset of myosin production contrasts with the interval seen in muscles developing in axial (approximately 18 hr) and wing (approximately 1-1.5 days) locations, and is unique to head muscles. This finding suggests that ongoing interactions between head myoblasts and their surroundings, most likely neural crest cells, delay myoblast withdrawal from the mitotic pool. These descriptions define a spatiotemporal pattern of muscle regulatory gene and myosin heavy chain expression unique to head muscles. This pattern is independent of origin (somitic vs. unsegmented paraxial vs. prechordal mesoderm), position (extraocular vs. branchial vs. subpharyngeal), and fiber type (fast vs. slow) and is shared among all muscles whose precursors interact with cephalic neural crest populations. Dev Dyn 1999;216:96-112.

Myotube heterogeneity in developing chick craniofacial skeletal muscles.

Avian skeletal muscles consist of myotubes that can be categorized according to contraction and fatigue properties, which are based largely on the types of myosins and metabolic enzymes present in the cells. Most mature muscles in the head are mixed, but they display a variety of ratios and distributions of fast and slow muscle cells. We examine the development of all head muscles in chick and quail embryos, using immunohistochemical assays that distinguish between fast and slow myosin heavy chain (MyHC) isoforms. Some muscles exhibit the mature spatial organization from the onset of primary myotube differentiation (e.g., jaw adductor complex). Many other muscles undergo substantial transformation during the transition from primary to secondary myogenesis, becoming mixed after having started as exclusively slow (e.g., oculorotatory, neck muscles) or fast (e.g., mandibular depressor) myotube populations. A few muscles are comprised exclusively of fast myotubes throughout their development and in the adult (e.g., the quail quadratus and pyramidalis muscles, chick stylohyoideus muscles). Most developing quail and chick head muscles exhibit identical fiber type composition; exceptions include the genioglossal (chick: initially slow, quail: mixed), quadratus and pyramidalis (chick: mixed, quail: fast), and stylohyoid (chick: fast, quail: mixed). The great diversity of spatial and temporal scenarios during myogenesis of head muscles exceeds that observed in the limbs and trunk, and these observations, coupled with the results of precursor mapping studies, make it unlikely that a lineage based model, in which individual myoblasts are restricted to fast or slow fates, is in operation. More likely, spatiotemporal patterning of muscle fiber types is coupled with the interactions that direct the movements of muscle precursors and subsequent segregation of individual muscles from common myogenic condensations. In the head, most of these events are facilitated by connective tissue precursors derived from the neural crest. Whether these influences act upon uncommitted, or biased but not restricted, myogenic mesenchymal cells remains to be tested.

Cell origins and tissue boundaries during outflow tract development.

Morphogenesis of the cardiac outflow tract and aortic sac regions requires the progressive immigration and integrated differentiation of cells having very divergent embryonic histories. Mesodermal cells originating both within and beside the developing head contribute to endocardium and myocardium. These cells, together with later arriving neural crest cells, participate in the formation of the aorticopulmonary septum, truncal cushions, and semilunar valves, although there is uncertainty regarding the precise contributions of each. In addition, precursors of the enveloping epicardium and coronary arteries move into the outflow tract. Defining the spatial and temporal contributions of these disparate populations and the boundaries between them as the outflow tract shifts caudally is an essential prerequisite to understanding normal heart morphogenesis as well as the etiology of outflow tract dysmorphologies.

Developmental relations between sixth nerve motor neurons and their targets in the chick embryo.

The developmental relations between abducens (VI) nerves and their targets, the lateral rectus, quadratus, and pyramidalis muscles, have been examined in the chick embryo from early neural tube stages through 10 days of incubation. Sites of myoblast origins were determined by microinjection of replication-incompetent retroviruses containing the LacZ reporter into paraxial mesoderm corresponding to somitomeres 3-5. Motor neurons and axons were identified by Bodian staining, immunocytochemistry, and application of DiI and DiO to dissected peripheral nerves. Anlage of the dorsal oblique originate in somitomere 3, close to the ventrolateral margin of the mid-to-caudal mesencephalon. Precursors of the lateral rectus arise deep within somitomere 4, beside the future metencephalon (rhombomere "A"). Quadratus and pyramidalis precursors are located between and partially segregated from these other two anlage. VIth nerve axons exit rhombomeres 5 and 6 via multiple median roots, fasciculate, and by stage 17 have elongated rostrally beneath the hindbrain. Immediately caudal to a mesenchymal pre-muscle condensation located deep to rhombomere 2, the VIth nerve separates into two branches. One branch enters the rostral portion of the condensation, from which quadratus and pyramidalis muscles will segregate. This branch projects exclusively from rhombomere 5 and is the accessory abducens nerve. The other branch enters the caudal, presumptive lateral rectus, region of the condensation. This is the abducens nerve, and it projects from cells located in both rhombomeres 5 and 6. These findings indicate that specific matching of motor nerves with their presumptive targets begins prior to the differentiation and segregation of myogenic populations, and that spatial organization of developing eye muscles is initiated well before they interact with connective tissue precursors derived from the neural crest.

Re-evaluating the role of astrocytes in blood-brain barrier induction.

Neural tissue induces brain capillary endothelial cells to express a diverse array of characteristics that allow them to regulate the passage of solutes between the blood and the brain; these features are collectively referred to as the blood-brain barrier (BBB). Because astrocytes are intimately associated with brain capillaries, they have been thought to be the cell type responsible for barrier induction. Widely accepted support of this hypothesis has been derived from experiments showing that astrocytes implanted into the anterior chamber of the rat eye, or onto the chorioallantoic membrane of the chicken embryo, remain unstained by circulating Evan's blue, while grafts of fibroblasts in these sites stain intensely. We have found several limitations associated with placing grafts in either site, leading us to believe that previously reported results are inconclusive. Astrocytes implanted into the anterior chamber form grafts that are poorly vascularized, whereas fibroblast grafts are richly vascularized by vessels which are often fenestrated. This likely accounts for apparent differences in vessel permeability reported by others. We have found that iridial vessels associated with astrocyte grafts do not change their ultrastructure to resemble brain capillaries. Grafting of cells to the chorioallantoic membrane elicits an extensive inflammatory response. Inflammation results in poor delivery of tracers to graft vasculature as well as altering vessel permeability. Treatment of hosts with steroidal anti-inflammatory agents in doses compatible with survival of the host does allow improved graft survival. Even after treatment with anti-inflammatory agents, however, astrocyte graft vasculature fails to express high levels of a barrier marker, the GLUT-1 isoform of the glucose transporter. Transplantation of avascular embryonic spinal cord, that induces robust vessel ingrowth and GLUT-1 expression in intra-embryonic vessels, was unable to elicit the ingrowth of more than a few vessels from the chorioallantoic membrane vasculature, and none of these expressed glucose transporter. We conclude that the anterior chamber and chorioallantoic membrane are not suitable sites for studying BBB induction, and that there is, at present, no conclusive evidence that mature astrocytes play a significant role in the initial expression of the BBB.

Spatial integration among cells forming the cranial peripheral nervous system.

Neural crest cells represent a unique link between axial and peripheral regions of the developing vertebrate head. Although their fates are well catalogued, the issue of their role in spatial organization is less certain. Recent data, particularly on patterns of expression of Hox genes in the hindbrain and crest cells, have raised anew the debate whether a segmental arrangement is the basis for positional specification of craniofacial epithelial and mesenchymal tissues or is but one manifestation of underlying spatial programming processes. The mechanisms of positional specification of sensory neurons derived from the neural crest and placodes are unknown. This review examines the spatial organization of cells and tissues that develop in proximity to sensory neurons; some of these tissues share a common ancestry, others are targets of cranial sensory and motor nerves. All share the necessity of acquiring and expressing site-specific properties in a functionally integrated manner. This integration occurs in part by coordinating patterns of cell migration, as occurs between migrating crest cells and branchial arch myoblasts. Constant rostro-caudal relations are maintained among these precursors as they move dorsoventrally from the hindbrain-paraxial regions to establish branchial arches. During this period the interactions among these and other mesenchymal cells are hierarchical; each cell population differentially integrates its past with cues emanating from new microenvironments. Analyses of tissue interactions indicate that neural crest cells play a dominant role in this scenario.

Vertebrate craniofacial development: novel approaches and new dilemmas.

Unexpected patterns of neuroblast, angioblast and myoblast movement and tissue organization have been defined using lineage tracing and transplantation methods. The most novel and enigmatic new data derive from analyses of genes in the Hox- and Pax-gene families. In addition to the characterization of expression patterns, the effects of Hox-gene knock-out and retinoic acid treatment have been assessed. These basic studies are complemented by the identification of correlations between inherited craniofacial anomalies, for example Waardenburg's syndrome, and the function of specific genes.

Studies on avian spinal cord chimeras. I. Neurological and histological manifestations.

Spinal cord chimeras were produced by replacing a small fragment of neural tube of a 2-day-old White Leghorn chicken embryo with a similar fragment from a Japanese quail embryo. The embryo mortality was 61%, and 72% of hatched birds were 'cripples' and had to be sacrificed within 5 days after hatching. Forty-nine chimeras, 10.9% of the total number of operated embryos, were alive for more than 3 weeks. For at least 17 days after hatching, all birds behaved like normal chicks, and the grey quail-like feathers were the only manifestations of their chimerism. Initial neurological symptoms of unsteady walking and drooping of the wings were noted in all birds except for 1 that died an accidental death before it became sick. Advanced symptoms characterized by paralysis of the legs forcing the bird to lie on its side were noted in 40 birds. The chimeras could be divided into two groups, each consisting of 24 birds. The short-survival (SS) chimeras of the first group became terminally ill and had to be sacrificed within 3 months. The long-survival (LS) chimeras of the second group showed more protracted disease, in that only 16 of them showed symptoms of the advanced disease, and the majority showed partial or complete recovery. Ten of the LS birds were kept alive for more than 8 months. Furthermore, many LS chimeras lost their grey feathers. The hallmarks of neurohistological manifestations were mononuclear cell infiltrates, demyelinization with preservation of axons and scar formation. These lesions were restricted to the quail fragment of the spinal cord except for 2 birds in which distant cellular infiltrates were observed. Direct immunofluorescence tests for chicken IgG were positive in spinal cords of most SS chimeras but only of some LS chimeras.

Cell movements and control of patterned tissue assembly during craniofacial development.

Craniofacial mesenchyme is heterogeneous with respect to origins (e.g., paraxial mesoderm, lateral mesoderm, prechordal mesoderm, neural crest, placodes) and fates. The many disparate cell migratory behaviors exhibited by these mesenchymal populations have only recently been revealed, necessitating a reappraisal of how these different populations come together to form specific tissues and organs. The objectives of this review are to characterize the diverse migratory behaviors of craniofacial mesenchymal subpopulations, to define the interactions necessary for their assembly into tissues, and to discuss these data in the context of recent discoveries concerning the molecular basis of craniofacial development. The application of antibodies that recognize features unique to migrating neural crest cells has verified the results of previous transplantation experiments in birds and shown the migratory pathways in murine embryos to be similar. Within paraxial or prechordal mesoderm arise myoblasts that are precursors of craniofacial voluntary muscles. These cells migrate, usually en masse, to the sites where overt muscle differentiation occurs. Whereas the initial alignment of primary myotubes presages the fiber orientation seen in the adult, the time at which individual myotubes appear relative to the formation of discrete, individual muscle bundles and attachments with connective tissues varies with each muscle. The pattern of primary myotube alignment is determined by local connective tissue-forming mesenchyme and is independent of the source of myoblasts. Also found within paraxial and lateral mesodermal tissues are endothelial precursors (angioblasts). Some of these aggregate in situ, forming vesicles that coalesce with ingrowing endothelial cords. Others are highly invasive, moving in all directions and infiltrating tissues such as the neural crest, which lacks endogenous angioblasts. The patterns of initial blood vessel formation in the head are also determined by local connective tissue-forming mesenchyme and are independent of the origin of endothelial cells. Neural crest cells, which constitute the predominant connective tissue-forming mesenchyme in the facial, oral, and branchial regions of the head, acquire a regional identity while still part of the neural epithelium, and carry this with them as they move into the mandibular, hyoid, and branchial arches. Some of these regionally unique propensities correspond spatially to genetic and cellular patterns unique to rhombomeres, although the links between gene expression and crest population phenotypes are not yet known. In contrast, the inherent spatial programming of those crest cells that populate the maxillary and frontonasal regions is altered by their proximity to the prosencephalon.

Experimental analysis of blood vessel development in the avian wing bud.

Shortly after its appearance, the avian limb bud becomes populated by a rich plexus of vascular channels. Formation of this plexus occurs by angiogenesis, specifically the ingrowth of branches from the dorsal aorta or cardinal veins, and by differentiation of endogenous angioblasts within limb mesoderm. However, mesenchyme located immediately beneath the surface ectoderm of the limb is devoid of patent blood vessels. The objective of this research is to ascertain whether peripheral limb mesoderm lacks angioblasts at all stages or becomes avascular secondarily during limb development. Grafts of core or peripheral wing mesoderm, identified by the presence or absence of patent channels following systemic infusion with ink, were grafted from quail embryos at stages 16-26 into the head region of chick embryos at stages 9-10. Hosts were fixed 3-5 days later and sections treated with antibodies that recognize quail endothelial cells and their precursors. Labeled endothelial cells were found intercalated into normal craniofacial blood vessels both nearby and distant from the site of implantation following grafting of limb core mesoderm from any stage. Identical results were obtained following grafting of limb peripheral mesoderm at stages 16-21. However, peripheral mesoderm from donors older than stage 22 did not contain endothelial precursors. Thus at the onset of appendicular development angioblasts are present throughout the mesoderm of the limb bud. During the fourth day of incubation, these cells are lost from peripheral mesoderm, either through emigration or degeneration.

Origins and patterning of avian outflow tract endocardium.

Outflow tract endocardium links the atrioventricular lining, which develops from cardiogenic plate mesoderm, with aortic arches, whose lining forms collectively from splanchnopleuric endothelial channels, local endothelial vesicles, and invasive angioblasts. At two discrete sites, outflow tract endocardial cells participate in morphogenetic events not within the repertoire of neighboring endocardium: they form mesenchymal precursors of endocardial cushions. The objectives of this research were to document the history of outflow tract endocardium in the avian embryo immediately prior to development of the heart, and to ascertain which, if any, aspects of this history are necessary to acquire cushion-forming potential. Paraxial and lateral mesodermal tissues from between somitomere 3 (midbrain level) and somite 5 were grafted from quail into chick embryos at 3-10 somite stages and, after 2-5 days incubation, survivors were fixed and sectioned. Tissues were stained with the Feulgen reaction to visualize the quail nuclear marker or with antibodies (monoclonal QH1 or polyclonals) that recognize quail but not chick cells. Many quail endothelial cells lose the characteristic nuclear heterochromatin marker, but they retain the species-specific epitope recognized by these antibodies. Precursors of outflow tract but not atrioventricular endocardium are present in cephalic paraxial and lateral mesoderm, with their greatest concentration at the level of the otic placode. Furthermore, the ventral movement of individual angiogenic cells is a normal antecedent to outflow tract formation. Cardiac myocytes were never derived from grafted head mesoderm. Thus, unlike the atrioventricular regions of the heart, outflow tract endocardial and myocardial precursors do not share a congruent embryonic history. The results of heterotopic transplantation, in which trunk paraxial or lateral mesoderm was grafted into the head, were identical, including the formation of cushion mesenchyme. This means that cushion positioning and inductive influences must operate locally within the developing heart tubes.

Vertebrate craniofacial development: the relation between ontogenetic process and morphological outcome.

Many structures that are present, often transiently, in the head of extant vertebrate embryos appear to be segmentally organized. These include the brain, particularly the hindbrain (e.g., rhombomeres), and adjacent axial structures such as paraxial mesoderm (e.g., somites, somitomeres) and neural crest cells. Also present in the head are additional sets of serially arranged structures that develop in more ventral and lateral locations. Examples of these are epibranchial placodes, aortic arches, and pharyngeal pouches. All these embryonic structures are frequently used both individually and collectively as characters to assist in defining homologies. New cell labeling and identification methods are providing detailed accounts of cell movements and tissue lineages that reveal a range of disparate behaviors not previously appreciated. The well-known migrations of neural crest cells bring all but the neurogenic members of this mesenchymal population form dorsal, axial locations into ventral and rostral locations where they largely surround the pharynx, stomdeum, and prosencephalon. Equally dramatic movements of neural plate cells, myoblasts, angioblasts, and placode-derived cells have recently been documented. These movements may occur in concert with those of other nearby tissues (e.g., branchiomeric myoblasts, neural crest cells, and surface ectoderm) or may be independent (e.g., placodal neuroblasts). Migrating cells may be clustered and follow definable pathways towards their destination (e.g., neural crest cells), or they may be solitary and wander invasively without a prespecified destination (e.g., angioblasts). These extensive morphogenetic movements bring cells into contact with a greater variety of other tissues and matrix environments than has heretofore been recognized. Moreover, because of these rearrangements, the cells present in a particular location, such as a branchial arch, may trace their ancestry to many axial levels, which complicates the analyses of segmental relations. Comparative morphological studies of craniofacial development have recently been augmented by descriptions of the sites and times of expression of many matrix components, growth factors and their receptors, and regulatory genes. Particularly important has been the discovery of a network of genes called the homeobox family. These genes are similar in their sequence and their organization along a chromosome to genes that establish the spatial identity of prospective body parts in drosophila. The combination of cellular and molecular descriptive studies of vertebrate craniofacial development provide exciting opportunities to catalogue patterns of gene expression and morphogenesis during the gastrula, neurula, and early organogenesis stages. Moreover, such data form the basis for proposing and then testing hypotheses about the mechanisms controlling cell movements, tissue formation, and the assembly of functionally integrated sets of structures.(ABSTRACT TRUNCATED AT 400 WORDS)

Origins and assembly of avian embryonic blood vessels.

Two processes by which embryonic blood vessels develop are well-known: angiogenesis (growth by budding and branching of existing vessels) and local formation of endothelial vesicles that coalesce with elongating vessels. The former process appears to be more prevalent, with the latter restricted to vessels that form near the endoderm-mesoderm interface. The contributions of endothelial cells formed by each of these processes to specific blood vessels has not been defined, however, nor have the origins of precursors (angioblasts) of intraembryonic endothelial populations been established. To identify the origins of endothelial cells, precursor populations from quail embryos were transplanted into chick embryos. Antibodies that recognize quail endothelial cells were applied to sections from chimeric embryos fixed 2-5 days after surgery. These experiments reveal that all intraembryonic mesodermal tissues, except the notochord and prechordal plate, contain angiogenic precursors. Many angioblasts emigrate from the grafted tissue, invading surrounding mesenchyme and contributing to the formation of arteries, veins, and capillaries in a wide area. The invasive behavior of these angioblasts is unlike that of any other embryonic mesenchymal cell type and represents a third process operating during embryonic blood vessel formation. Transplanted angioblasts, even those excised from quail trunk regions, form normal craniofacial vascular channels, including the cardiac outflow tract. These results demonstrate that the control over blood vessel assembly resides within the connective tissue-forming mesenchyme of the embryo, not within endothelial precursors.

Embryonic origins and assembly of blood vessels.

Embryonic blood vessels develop in two ways: angiogenesis, which is growth by budding, branching, and elongation of existing vessels, and in situ formation of endothelial vesicles that coalesce with elongating vessels. It is assumed that the former is more prevalent, with the latter restricted to vessels that form near the endoderm:mesoderm interface. Neither the relative contributions of each of these processes in the formation of specific blood vessels nor the origins of precursors (angioblasts) of these intraembryonic endothelial populations are known. Antibodies that recognize quail endothelial cells can be used to follow the movements and differentiation of endothelial cell precursors after the transplantation of putative precursor populations from quail into chick embryos. Using this method, it has been shown that all intraembryonic mesodermal tissues, except the prechordal plate, contain angiogenic precursors. After transplantation some angioblasts move in all directions away from the site of implantation, invading surrounding mesenchyme and contributing to the formation of arteries, veins, and capillaries in a wide area. Although it is clear that these invasive angioblasts, which behave unlike any other embryonic mesenchymal cell type, are found throughout the embryo, it is not known whether they represent a unique endothelial cell type in mature blood vessels. Irrespective of their original location in the donor embryo, transplanted angioblasts will form vascular channels that are appropriate for the tissues surrounding their site of implantation. These results indicate that the control over vascular assembly resides within the connective-tissue-forming mesenchyme of the embryo.

SEM characterization of a cellular layer separating blood vessels from endoderm in the quail embryo.

The associations between the developing blood vessels and both endoderm and splanchnic mesoderm in quail embryos at stages 9-11 were examined by using scanning electron microscopy. Embryos were pinned ventral-side up on agar plates and the endoderm was surgically removed prior to fixation and dehydration. This procedure exposes a netlike layer of cells closely apposed to the ventral surface of paraxial mesoderm and all visible blood vessels; we are calling this the subvascular layer. Development of this layer proceeds rostral-to-caudal, and lateral-to-medial, with the earliest stages of formation being visible over the unsegmented paraxial mesoderm of the segmental plate. The subvascular layer increases markedly in density slightly medial to the innermost boundary of the intraembryonic vascular plexus. Cells of this layer eventually establish a continuous sheet beneath the lateral plate and paraxial mesoderm and the notochord. With maturation, the cells of the subvascular layer approach confluence. The spatial and temporal patterns of development of the embryonic vascular tissues and the subvascular layer are closely correlated, suggesting a possible role for the subvascular layer in normal embryonic vascular development.