Cephalic neurulation in the mouse embryo analyzed by SEM and morphometry

A detailed account of mouse neurulation is given based mostly on SEM analysis over 20 hr of development. Many observations and measurements were made on staged living embryos and on embryos prepared for scanning and light microscopy to help deduce what mechanisms may contribute to neural tube formation. Each-lateral half of the early cephalic neural plate makes a convex bulge, opposite to the way it must fold to form a tube. Underlying mesenchyme and matrix are reported to have a role in forming these bulges. Processes that form the tube must overcome this opposed folding and the forces that produce it. Cranial flexure begins long before tube formation. The flexure commences at the rostral tip of the cephalic neural plate, then the apex of the flexure migrates caudally to the mesencephalic region. Early appearance of this flexure imposes a mechanical impediment to tube closure in forebrain and midbrain regions. Tube closure begins in the cervical region exactly where the neural plate is reflected dorsally by a bend in the embryo. This bend may mechanically assist closure in this region. Cells of the mouse neural plate are reported to contain organized microfilaments and microtubules, and the plate cells appear to change shape (reduce apical area and increase cell height) in the same manner as that suggested in embryos of some other species to contribute to neural tube formation. Measurements show that the lateral edges of the cephalic neural plate elongate craniocaudally more than the midline of the plate through each period. This elongation could contribute to the folding of the plate into a tube. The progress of cranial ventral flexure pauses while tube formation occurs, but edge elongation continues, presumably contributing to tube formation. There is considerable increase in volume of the neural plate during tube closure, and cell proliferation and enlargement of daughter cells seem sufficient to account for this growth. Mitotic spindles are positioned to place the majority of the daughter cells into the long axis of the neural plate, so ordered growth may be the main mechanism of elongation of the plate in the craniocaudal direction, which in turn may assist in tube formation. Mouse cephalic neural plates appear overlying already segmented cranial mesenchyme according to previous reports, and neuromeres develop precociously in the open plates, where their positions correlate exactly with the underlying segmented mesenchyme.     

The formation of the embryonic mesoderm in the early post-implantation mouse embryo

Summary The formation of the mesoderm in early post-implantation mouse embryos is described and analysed. The outgrowth of the mesoderm was found to depend on the changes in the shape of the embryonic ectoderm, which lead to a relative displacement of the primitive streak in the caudal direction. The primitive streak deposits its cells laterally in the case of the lateral mesoderm, and medially in the case of the headprocess. In doing so, the primitive streak leaves a trail of mesoderm cells. This means that mesoderm cells do not migrate actively from the caudally located primitive streak towards more frontal positions in the embryo. This is confirmed by the results of scanning electron microscopy, which revealed that mesoderm cells show no polarity at all in the caudofrontal direction. In may therefore be concluded that these cells probably do not migrate.Studies on the cell-cycle parameters of the embryonic ectoderm, showed that mesoderm cells-to-be are probably recruited not only from the proliferation zone, but also from the lateral ectoderm. It is oostulated that the lateral ectoderm gives rise, via the largest part of the primitive streak, to most of the mesoderm cells, whereas the proliferation zone gives rise to the head-process mesoderm, via the anterior part of the primitive streak.     

Ultrastructure of secondary neurulation in the chick embryo

Formation of the future lumbosacral level of the spinal cord was studied in two-day-old chick embryos by light and electron (transmission and scanning) microscopy. A neurulation overlap zone occupied this level. The dorsal portion of the neural tube formed by bending of the neural plate and approximation and fusion of neural folds (i.e., by primary neurulation), and the ventral part formed during secondary neurulation by cavitation of an initially solid, compact mass of cells, the medullary cord, derived from the tail bud. Secondary neurulation involved four morphogenetic processes: (1) segregation of the cells of the prospective medullary cord from cells of adjacent regions, (2) formation of a precisely delimited medullary cord, (3) cavitation of the central portion of this cord, and (4) coalescence of all lumina into a single, central cavity. Cell segregation was associated with the formation of a layer of primarily extracellular materials between adjacent organ rudiments. The source and composition of these materials are unknown. Formation of the medullary cord entailed considerable elongation of the peripheral cells of this developing structure and the fabrication of small intercellular junctions, first at the basal (outer) ends of the elongating peripheral cells, and then at their apical (inner) ends. These events resulted in the formation of an outer pseudostratified layer of radially arranged, columnar cells, having characteristics similar to those of the neural plate, and an inner cluster of irregularly shaped and arranged cells. Cavitation always occurred first at the junction between these two cellular populations. The central cells of the medullary cord also eventually elongated, like the peripheral cells, and may have been intercalated into the lateral walls of the developing neural tube as lumina coalesced.     

Distribution of glycoconjugates in the optic vesicle and optic cup

Summary Lectin histochemical methods and immunohistochemical techniques have been utilized to investigate and partially characterize glycoconjugates in the developing eye. Peanut-lectin-binding sites associated with radial glial cells were found in the diencephalon. In the optic primordia, binding sites associated with radial glia were masked by terminal sialic acid, and only reacted with peanut lectin when pretreated with sialidase. This finding indicates that glycoconjugates associated with diencephalic radial glia contain terminal galactose--(13)N-acetyl galactosamine, but glycoconjugates associated with radial glia in the optic primordia contain sialic acidgalactose-(13)N-acetyl galactosamine. The selective distribution of galactose, N-acetyl galactosamine and fucose associated with radial glial cells has also been demonstrated. We postulate that these distributions mediate the shaping of the developing eye.This work was supported by NIH Grants # NS 11066 and # DE 05832     

Vimentin intermediate filament protein as differentiation marker of optic vesicle epithelium in the chick embryo

For the study of the differentiation process of optic vesicle epithelium into neural retina, pigment epithelium and pars caeca retinae, vimentin intermediate filament protein in retinal epithelial cells was detected immunohistochemically in chick embryo at stages 11-21. In the late stage of optic vesicle development (stage 14), optic vesicle epithelium was classified into the following 3 different portions on the basis of vimentin staining intensity: latero-central epithelium under the lens placode, medio-central epithelium facing the latero-central epithelium, and peripheral epithelium connecting the latero-central and medio-central epithelia. Latero-central epithelium, the future neural retina, exhibited strongest staining of vimentin of the 3 portions. In contrast, medio-central epithelium, the future pigment epithelium, showed weakest staining. Moderate staining was observed in peripheral epithelium, the future pars caeca retinae. These differences in levels of vimentin expression were observed during optic cup formation. The present results clearly demonstrate that differentiation of retinal epithelium into neural retina, pigment epithelium and pars caeca retinae occurs in the late stage of the optic vesicle, and that retinal differentiation is reflected by the amount of vimentin in epithelial cells.     

Prostacyclin receptor signaling and early embryo development in the mouse

BACKGROUND: Prostacyclin (PGI(2)) plays an important role in mouse embryo development and implantation. However, it is unclear whether its action is mediated via the I prostaglandin receptor (IP). METHODS: We compared the preimplantation development of IP deleted (IP-/-) embryos and wild-type (WT) embryos. We also evaluated the effect of iloprost, a stable PGI(2) analog, and L-165041, a peroxisome proliferator activated receptor delta (PPARdelta) ligand, on IP-/- versus WT embryos. Finally, we compared the development of heterozygous IP deficient embryos carrying a normal maternal IP allele versus paternal IP allele. RESULTS: Development of IP-/- embryos lagged behind WT embryos and was not enhanced by either the PGI(2) analog or the PPARdelta ligand. WT embryos had slightly higher, although statistically not significant, implantation rates than IP-/- embryos. Heterozygous IP deficient embryos carrying a normal maternal IP allele showed better development and responded to the PGI(2) analog, unlike those carrying the normal paternal IP allele. CONCLUSIONS: IP receptors play an important role in preimplantation embryo development and mediate the embryo's response to exogenous PGI(2). Early embryo development depends on the oocyte IP receptor.     

Development of Reichert's membrane in the early mouse embryo

Although the composition of Reichert's membrane, a thick multilayered basement membrane between the parietal endoderm cells and the trophoblast cells of rodents, has often been investigated, the site of its production remains a subject of controversial discussion. In particular, the role of the trophoblast cells is unclear. In the present work we examined the initial development of Reichert's membrane in the early mouse embryo, using glutaraldehyde fixation with tannic acid. In the early blastocyst the occurrence of a tannic-acid-positive layer located at the inner surface of the mural trophoblast indicated the onset of basement membrane formation by the trophoblast cells. In the peri-implantation phase, this basement membrane extended into lateral areas of the inner cell mass separating the newly differentiated ectoderm and endoderm cells from each other. In these lateral regions, where the recently formed primitive endoderm cells had been attached to the monolayered basement membrane of the mural trophoblast, the membrane began to reveal the typical multilayered structure of Reichert's membrane. Our findings indicate that the initial formation of Reichert's membrane begins with the formation of a basement membrane of the mural trophoblast cells, followed by an apposition of basement membrane material, probably synthesized by primitive endoderm cells, along this primary membrane.     

Macrophages during retina and optic nerve development in the mouse embryo: relationship to cell death and optic fibres

We compared the spatial and temporal patterns of distribution of macrophages, with patterns of naturally occurring cell death and optic fibre growth during early retina and optic nerve development, in the mouse. We used embryos between day 10 of embryogenesis (E10; before the first optic fibres are generated in the retina) and E13 (when the first optic fibres have crossed the chiasmatic anlage). The macrophages and optic axons were identified by immunocytochemistry, and the apoptotic cells were detected by the TUNEL technique, which specifically labels fragmented DNA. Cell death was observed in the retina and the optic stalk long before the first optic axons appeared in either region. Subsequently, specialized F4/80-positive phagocytes were detected in chronological and topographical coincidence with cell death, which disappeared progressively. As development proceeded, the pioneer ganglion cell axons reached the regions where the macrophages were located. As the number of optic fibres increased, the macrophages disappeared. Therefore, cell death, accompanied by macrophages, preceded the growth of fibres in the retina and the optic nerve. Moreover, these macrophages synthesized NGF and the optic axons were p75 neurotrophin receptor (p75^NTR)- and TrkA-positive. These findings suggest that macrophages may be involved in optic axon guidance and fasciculation.     

The head-process and the formation of the definitive endoderm in the mouse embryo

Summary The formation of the secondary or definitive endoderm was studied by light microscopy (1-m sections) and (scanning) electron microscopy. The results show that the primary endoderm disappears axially, and a hiatus appears in this layer. The development of this hiatus may be caused by cell degeneration, which is observed in the primary endoderm, or by some activity of the underlying head-process. The apical parts of a number of head-process cells converge towards a hiatus. These cells are organized into a conical configuration which may participate in the formation of the hiatus. The cone cells reach through the hiatus into the yolk sac cavity, and comprise the secondary endoderm. The consequence is that in mice, the definitive endoderm develops from the head-process mesoderm rather than from the primary endoderm.     

A reexamination of the role of microfilaments in neurulation in the chick embryo

Formation of wedge-shaped neuroepithelial cells, owing to the constriction of apical bands of microfilaments, is widely believed to play a major part in bending of the neural plate. Although cell "wedging" occurs during neurulation, its exact role in bending is unknown. Likewise, although microfilament bands occupy the apices of neuroepithelial cells, whether these structures are required for cell wedging is unknown. Finally, although it is known that cytochalasins interfere with neurulation, it is unknown whether they block shaping or furrowing of the neural plate, or elevation, convergence, or fusion of the neural folds. The purpose of this study was to reexamine the role of microfilaments in neurulation in the chick embryo. Embryos were treated with cytochalasin D (CD) to depolymerize microfilaments and were analyzed 4-24 hr later. CD did not prevent neural plate shaping, median neural plate furrowing, wedging of median neuroepithelial cells, or neural fold elevation. However, dorsolateral neural plate furrowing, wedging of dorsolateral neuroepithelial cells, and convergence of the neural folds were blocked frequently by CD. In addition, neural folds always failed to fuse across the midline in embryos treated with CD, and neural crest cell migration was prevented. These data indicate that only the later aspects of neurulation may require microfilaments, and that certain neuroepithelial cells, particularly those that normally wedge with median furrowing and elevation of the neural folds, become (and remain) wedge-shaped in the absence of apical microfilament bands. Thus, microfilament-mediated constriction of neuroepithelial cell apices is not the major force for median neuroepithelial cell wedging and elevation of the chick neural plate. Further studies are needed to localize the motor(s) for these processes.     

Mesenchyme formation from the trigeminal placodes of the mouse embryo

The trigeminal placode is a thickened region of ectodermal epithelium located along the side of the embryonic head. Mesenchyme escapes from the placode to form neurons of the trigeminal (V) ganglion. To further our knowledge of the morphogenesis of this escape, plastic thick sections were cut from mouse embryos and stained for light microscopy by using a technique which revealed escaping mesenchyme. The escape of trigeminal mesenchyme began at approximately 12 somites of age and was substantially complete by 30 somites. These results provided spatial/temporal orientation for a subsequent electron microscopic study. The first ultrastructural manifestation of escape was the penetration of an otherwise continuous basal lamina by small cell processes. The presence of longitudinally oriented microtubules within these processes suggests that mesenchymal cells escape through the basal lamina by using microtubules to direct/move their contents (e.g., the cell nucleus) into an enlarging process. Nuclei were distorted as they passed into these processes. This distortion suggests that basal lamina, together with a possible contribution from basal microfilaments, forms a rigid obstruction which is disrupted in the region from which a process is formed. In some cases a collar of basal lamina was observed around the necks of processes, but their distal membranes were invariably lamina-free. This lamina-free membrane is possibly that which is newly formed to accommodate the growing process. In later stages of escape, instances were observed in which the lamina was completely absent beneath an escaping cell and partially degraded beneath adjacent cells as well. These instances suggest that enzymatic digestion may play a role in degrading the lamina during mesenchymal escape. Apical desmosomes were often retained beyond the initial stages of escape. Mechanisms involved in their disruption are thus not among those which initiate escape.     

Development of pharyngeal arch arteries in early mouse embryo

The formation and transformation of the pharyngeal arch arteries in the mouse embryo, from 8.5 to 13 days of gestation (DG), was observed using scanning electron microscopy of vascular casts and graphic reconstruction of 1-microm serial epoxy-resin sections. Late in 8.5-9DG (12 somites), the paired ventral aortae were connected to the dorsal aortae via a loop anterior to the foregut which we call the 'primitive aortic arch', as in the chick embryo. The primitive aortic arch extended cranio-caudally to be transformed into the primitive internal carotid artery, which in turn gave rise to the primitive maxillary artery and the arteries supplying the brain. The second pharyngeal arch artery (PAA) appeared late in 9-9.5DG (16-17 somites), and the ventral aorta bent dorsolaterally to form the first PAA anterior to the first pharyngeal pouch by early in 9.5-10DG (21-23 somites). The third PAA appeared early in 9.5-10DG (21-23 somites), the fourth late in 9.5-10DG (27-29 somites), and the sixth at 10DG (31-34 somites). By 10.5DG (35-39 somites), the first and second PAAs had been transformed into other arteries, and the third, fourth and sixth PAAs had developed well, though the PAA system still exhibited bilateral symmetry. By 13DG, the right sixth PAA had disappeared, and the remaining PAAs formed an aortic-arch system that was almost of the adult type.     

Deceleration and acceleration in the rate of posterior neuropore closure during neurulation in the curly tail (ct) mouse embryo

Summary Curly tail (ct) is a mouse mutant producing spinal neural tube defects as a result of delayed closure of the posterior neuropore (PNP). The purpose of the present study was to determine in ct/ct embryos the time of onset of the delay in PNP closure, and the pattern of this closure, as well as to study the possibility that reopening of the neural tube occurs. Normal spinal neurulation was studied in non-mutant Swiss (Sw) embryos. In the latter, the average PNP length diminished steadily between the 7- and 25-somite stages, and then decreased more rapidly, indicating an acceleration of closure rate, until the 30- to 32-somite stage, when all PNPs closed. PNP width decreased steadily between the stages of 7 and 30 somites. In ct/ct embryos the average PNP length showed a slight increase between the stage of 23 to 28 somites, indicating a temporary deceleration of closure rate, and the range of PNP sizes increased markedly. This was followed by a decrease in PNP length until the 37-somite stage, indicating an acceleration of closure rate. From the stage of 32 somites onwards, the proportion of embryos with closed PNPs gradually increased to 90%. The population of ct/ct embryos was subdivided. Embryos with large PNPs showed a marked deceleration of closure rate during a period of 11 somite stages, followed by a brief but very high acceleration of closure rate. This resulted in closure of the PNP in a proportion of these embryos, while in the remainder of the embryos the deceleration phase had been too enhanced to allow complete catch up of closure during the acceleration phase; these embryos would develop spina bifida. Embryos with relative small PNPs also showed a deceleration of closure rate, but only during a period of four somite stages. This was followed by an acceleration, resulting in closure of all PNPs at the stage of 32 to 33 somites. The enlargement of the PNP in ct/ct embryos was not due to re-opening of a closed neural tube, but resulted from a sharp decline in the rate of PNP closure combined with a normal rate of caudal elongation of the embryo. It is concluded that the ct strain forms a homogeneous population, with a large variation of its specific phenotype: deceleration of PNP closure during a restricted period. The disturbance of spinal neurulation in ct/ct embryos takes the form of a deceleration/acceleration pattern, resulting in a net delay of closure. It is suggested that, due to the ct mutation, forces are generated in the embryonic axis which oppose a normal neurulation process at a specific stage of development.     

Distribution of cell surface glycoconjugates during secondary neurulation in the chick embryo

Lectin histochemistry was used to examine the expression of cell surface glycoconjugates during secondary neurulation in chick embryos. Fourteen lectins were applied to serial sections of the caudal region of embryos at the various stages of tail bud development. The lectins Bandeiraea simplicifolia, Dolichos biflorus agglutinin, Phaseolus vulgaris leukoagglutinin, soybean agglutinin, Sophora japonica agglutinin, Ulex europaeus agglutinin and succinylated wheat germ agglutinin (sWGA) showed very light or no binding to the developing medullary cord of the tail bud. With the other lectins, staining occurred throughout the early tail bud and solid medullary cord. During cavitation, however, differential expression of cell surface glycoconjugates by different cell populations was observed. The lectins concanavalin A, Lens culinaris agglutinin, Pisum sativum agglutinin, Phaseolus vulgaris erythroagglutinin, Ricinus communis agglutinin and WGA showed basic similarities in the distribution of lectin binding. Of these, the binding pattern of WGA was the most striking. As the medullary cord cells were separating into central mesenchymal and peripheral epithelial populations, WGA bound preferentially to the epithelial cells and the notochord. The lectin PNA, however, became preferentially bound to the mesenchymal cells. Heavy staining by WGA (specific for N-acetylglucosamine and sialic acid) where sWGA staining (specific for N-acetylglucosamine only) was faint suggested that WGA binding was due to the presence of sialic acid containing glycoconjugates.     

Quantitative studies of mitotic cells in the chick embryo optic stalk during the early period of invasion by optic fibres

Summary In addition to mitoses of neuroepithelial cells at the ventricular surface of the chick embryo optic stalk, mitoses in nonventricular stalk zones begin to be observed from stage 19 on. These latter represent the division phase of glioblasts detached from the ventricular surface. Thus, the topographical location of mitotic cells could be considered a morphological marker of neuroepithelial and glioblast populations in the optic stalk. Quantitative analysis of ventricular (VMCs) and extraventricular (EMCs) mitotic cells revealed that the total number of VMCs decreases through the developmental stages studied, while the number of EMCs simultaneously increases exponentially. These results suggest that the glioblast population arises from both division of the early glioblasts and progressive transformation of neuroepithelial cells.The first EMCs in the ventral region of the stalk wall are observed in stage 19, previous to the stages in which the first EMCs appear in the dorsal region. Moreover, EMCs are much more numerous in the ventral than in the dorsal stalk wall in all stages analysed. Keeping in mind that the invasion of the stalk by optic fibre fascicles occurs essentially in the ventral region, these results suggest that EMCs are strongly related to axon fascicle outgrowth in the stalk.Cell division features are different in neuroepithelial cell and glioblast populations, as the proportions of the mitotic phases differ in VMCs and EMCs. In addition, the patterns of mitotic spindle orientation in VMCs and EMCs are also different. In the former, orientations are predominantly longitudinal parallel and transverse parallel, with a smaller proportion of radial mitoses, which are slightly more frequent through stages 23 to 28 than in earlier development. In the EMCs, radial and longitudinal parallel spindle orientations are the most frequent, the proportion of mitoses with transverse parallel orientation being very low through stages 24 to 28. The significance of these results is discussed with reference to stalk developmental mechanisms.     

Histological and ultrastructural studies of secondary neurulation in mouse embryos

The histological and ultrastructural features of secondary neurulation in C57BL/6 mouse embryos were examined as a first step in the analysis of how this process occurs in mammalian embryos. Secondary neurulation involves two major events in mouse embryos: (1) formation of the medullary rosette (9.5- to 10-day embryos) or plate (11- to 12-day embryos), and (2) cavitation. These two events occur simultaneously. The medullary rosette consists of elongated tail bud cells, radially arranged around a central lumen formed by cavitation. The secondary portion of the neural tube forms in 9.5- to 10-day embryos by progressive enlargement of the central lumen and addition (by cell recruitment or mitosis) of tail bud cells to the rosette. The medullary plate likewise consists of elongated tail bud cells, but these cells do not surround a central cavity. Instead, cells of the medullary plate extend ventrad from the basal aspect of the dorsal surface ectoderm to a slit-like cavity formed by cavitation. Formation of the secondary neural tube occurs in 11- to 12-day embryos, principally by the recruitment of more lateral and ventral tail bud cells into the medullary plate. Free cells and cellular debris are frequently encountered in the forming lumen of the secondary neural tube, but cells exhibiting signs of necrosis were absent in cavitating regions. Numerous small intercellular junctions form at the inner (juxtaluminal) ends of tail bud cells as the medullary rosette or plate is forming and cavitation is occurring. These observations suggest that cavitation per se (i.e., formation of a lumen) during secondary neurulation is a relatively passive phenomenon, which results principally from neighboring cells becoming polarized apicobasally and incorporated into a primitive neuroepithelium. The latter constitutes the walls of the forming secondary neural tube.