Mapping of the early neural primordium in quail-chick chimeras

Mapping of the early neural primordium in quail-chick chimeras

DEVELOPMENTAL BIOLOGY 120,1%-214 (1987) Mapping of the Early Neural Primordium in Quail-Chick Chimeras II. The Prosencephalic Neural Plate and Neu...

47MB Sizes 0 Downloads 51 Views

DEVELOPMENTAL

BIOLOGY 120,1%-214

(1987)

Mapping of the Early Neural Primordium in Quail-Chick Chimeras II. The Prosencephalic

Neural Plate and Neural Folds: Implications for the Genesis of Cephalic Human Congenital Abnormalities GERARD

In&it& d’Embq&gie

F. COULY AND NICOLE M. LE DOUARIN

du CNRS et du CollGgede France, .49bis, Avenue de la Belle-Gabrielle,

94180Nogent-sur-Marne, France

Received June 19,1986; accepted in revised form October1, 1986 Mapping of the avian neural primordium was carried out at the early somitic stages by substituting definite regions of the chick embryo by their quail counterpart. The quail nuclear marker made it possible to identify precisely the derivatives of the grafted areas within the chimeric cephalic structures. A fate map of the prosencephalic neural plate and neural folds is presented. Moreover the origin of the forebrain meninges from the pro- and mesencephalic neural crest is demonstrated. In the light of the data resulting from these experiments, we present a rationale for the genesis of malformations of the face and brain and of congenital endocrine abnormalities occurring in man. o 1987 Academic Press. Inc.

INTRODUCTION

We have undertaken the mapping of the neural primordium at the early neurula stage by the isotopic and isochronic substitution of defined ectodermal territories between quail and chick embryos. The morphogenetic movements by means of which the head and brain regions become finally positioned can thus be followed using the quail nuclear marker, since it labels definite areas that can be visualized at progressively older developmental stages. It the first article of this series (Couly and Le Douarin, 1985), we reported findings concerning the fate of the rostrolateral neural fold and adjacent neural plate territories. This enabled us to define specific regions from which functionally related peripheral and central structures are derived. Such was the case for the territory forming the adenohypophysis (primarily located in the rostra1 neural fold) and the hypothalamus lying just behind in the neural plate. Similarly, we delimited an area situated in the rostrolateral neural fold that forms the olfactory placode, and an adjacent neural plate area that forms the olfactory nerve and bulb. This continuous anlage subsequently splits up into a peripheral and a central organ. The other, most unexpected result from these experiments was the finding that the neural ridge contributes to the superficial ectoderm of the roof of the mouth, the olfactory cavities, and the face. It also appeared that the anteriormost part of the adult brain, the telencephalon, was not the most rostrally located anlage of the neural plate, this position being occupied by the presumptive diencephalon. We decided to pursue our investigations to establish the relative positions in the neural plate of the presumptive telencephalic and dien0012-1606/87 $3.00 Copyright Q 1987 by Academic Press, Inc. All rights of reproduction in any form reserved.

cephalic territories at these early stages and hence to infer their relative movements during cephalogenesis. Since the rostra1 neural ridge produces epithelial structures (placodes and superficial ectoderm), a precise analysis of the rostra1 limits of the neural crest territory was carried out. This led to an elucidation of the levels of the neural folds that are implicated in the massive migration of mesectodermal cells to the nasofrontal and maxillary regions. It was also seen that meninges covering the forebrain originate from the mesencephalic neural crest. From these observations we propose a rationale for the genesis of certain human congenital dysplasias such as the Sturge-Weber syndrome and of certain prosencephalic malformations associated with facial anomalies. MATERIAL

AND

METHODS

Material

Quail and chick eggs from commercial sources were used throughout these experiments. The stages of the embryos were determined either according to the development time tables of Hamburger and Hamilton (1951) for the chick and of Zacchei (1961) for the quail or, for the early stages of development, by referring to the number of pairs of somites formed. Microsurge?y The embryos were operated at three- and four-somite stages. The operation consisted of the removal of defined regions of the neural ridges and neural plate in chick 198

COULY AND LE DOUARIN

Avian

Neural

Primcndium

199

chicken and replaced by their quail counterparts at three rostrocaudal levels over a length of about 150 pm on the right or left sides (Fig. 8a). In IIIA, zone A (involving both neural ridge and adjacent neural plate) indicated in Table 1 was involved in the operation. In IIIB, the territory located caudally with respect to the region involved in experiment 3A was subjected to substitution (area B of Table 1). In IIIC, the area of ridge and plate indicated as C in Table 1 was the object of the experiment. Experiment I. Consisted in the replacement of the Experiment IV: Figure 12a shows the extent of the whole prosencephalic region of embryos by their quail graft which corresponds to the rostra1 neural ridge (see equivalent as indicated in Fig. 3a (see Table 1.) experiment I of our previous article, Couly and Le Experiment II. In this experiment the same area of Douarin, 1985) from which the adenohypophysis develthe prosencephalon as in Experiment I was involved in ops, plus the mediodorsal territory of the neural plate the operation but the excision was unilateral and cor- over a length of about 100 pm (see Table 1). responded to the right or left hemiforebrain only (Fig. Experiment I! This experiment involved a double 6a) (see Table 1). graft. The two grafted areas were: the rostra1 neural Experiment III. The rostrolateral neural ridge and a ridge and the same territory as in experiment IIIB portion of the adjacent neural plate was removed in (Fig. 13a).

embryos and their replacement by their quail counterparts taken from embryos at the same developmental stage. As mentioned in our previous article of this series (Couly and Le Douarin, 1985), quail and chick embryos develop at a slightly different rate, the quail reaching a given stage about 2 hr before the chick at these early times. The surgical procedure consists in cutting with a steel needle the regions to be either removed or grafted as described in Le Douarin (1982). The following series of experiments were performed:

FIG. 1. Chimeric embryos E5.5 in which the prosencephalic area has been replaced by its quail counterpart at the 3-somite stage. The overall morphology of the head is normal. The boundary between quail and chick territories is not visible (Bar = 1 mm).

200

DEVELOPMENTALBIOLOGY

VOLUME120,1087

Experiment VI. The rostra1 mesencephalic neural ridge caudal to zone C was involved (see Table 1). Examination

of Chimeras

The chimeric embryos were sacrificed at E2.5,3.5,4.5, 5.5, and 8.5 (stages 17 to 35 of Hamburger and Hamilton), embedded in wax, and cut into 5-pm serial sections. Subsequent staining by the Feulgen-Rossenbeck technique allowed quail and chick cells to be distinguished. RESULTS

Origin of the Prosencephukm and Associated Nonneural Structures from the Rostra1 Neural Plate and Neural Ridge Experiments were performed with a view to elucidating the relationships existing during ontogeny between the territories of the neural primordium yielding the forebrain and those giving rise to various non-neural structures. Experiment I. Two embryos were sacrificed at E3.5 and five at E5.5. The overall morphology of the head in these chimeras was normal (Fig. 1). Quail cells were present in the surface ectoderm over the frontal and rostra1 cranial region, the nasofrontal area, the rostra1 region of the maxillary bud, and the eyes, as indicated in Figs. 2a, b and Fig. 3. In the stomodeum, the epithelium of the roof and of the future nasal cavity was made up of quail cells (Figs. 3b, c, e). The underlying mesenchyme of the facial region corresponding to the premaxillary and maxillary bones and to the corresponding dermis were of chick host origin (Figs. 3b-e). In the brain, the entire telencephalon and diencephalon were derived from the graft (Figs. 2e, f; 4). The quail-chick epithelial transition was regularly located around the mesencephalon-diencephalon boundaries (Fig. 4e). As expected from the results of experiment I described in Couly and Le Douarin (1985), in which the rostra1 neural ridge was grafted, the adenohyphyseal glandular cords were of graft origin while the interstitial adenohypophyseal connective tissue was of the chick type (Fig. 4f). The neurohypophysis (Fig 4c), epiphysis, and the roof of the diencephalon were made up of quail cells (Fig. 4d). The optic vesicle were also graft derived. Periocular structures, including the cornea1 endothelium, the sclera, and the choroid are all derived from the neural crest. Some contribution to these tissues was found (Figs. 5a-c). Some quail mesenchymal cells were dispersed in the primordia of the meninges at the level of mid- and pos-

.‘..: ,’ .‘..,.,, ;.,.;;...‘,: .... ..‘...,. ‘0,’ 9 C

FIG. 2. Schematic drawings representing the extension of quail labeling in experiment I. In order: the epkkrmti (a, b) the mesectodermalmcsenchyme (c, d), which is chimeric in the frontal, ophthalmic, and rostra1 facial areas, and brain (e, f). Telencephalon and diencephalonderived structures are made up from the grafted territory. The hatched areas are entirely derived from the graft. In the spotted region, the mesenchyme is chimeric.

terior telencephalon and around the rostra1 and dorsal region of the optic vesicle in the future choroid and sclerotic layers (Figs. 2c, d; 5b). The mesenchyme that will later give rise to the dermis and calvaria, was also chimerit (Fig. 5b). Experiment II. Five embryos were observed (Table 1). This series involved the hemiterritory implanted in experiment I (Fig. 6a) and yielded essentially normal embryos (Figs. 6b, c). Note in Fig. 6c the smaller volume of the quail half as compared to the chick half prosencephalic region at E3.5 in the host. This discrepancy is transitory and the external gross anatomy of the face had a normal appearance from E5.5 onward in most cases. At E5.5 the hypothalamus (Fig. 6d) was made up

COULY

AND

LE DOUARIN

Avian Neural Primordium

201

FIG. 3. Illustration of the results summarized in Fig. 2. (a) Scanning electron micrograph of the rostra1 region of a chick embryo at the 3somite stage showing the territory involved in experiment I. (b) Transverse section of the embryo represented on Fig. 1 at the level of the nasal region in which the pictures represented in detail in (c) to (e) are positioned (Bar = 100 pm). (c) The nasal bud ectoderm is of the quail donor type, the mesenchyme is of the chick host type (Bar = 10 pm). (d) The maxillary bud ectoderm is of quail and the mesenchyme of chick type (Bar = 10 pm). (e) The olfactory placode is entirely of quail donor type (Bar = 10 pm).

202

DEVELOPMENTAL BIOLOGY VOLUME120,1987

COULY AND LE DOUARIN

Avian

Neural

Primodium

203

b

FIG. 5. Detailed views of the left eye of the embryo represented in Fig. 1. The lens (L) (a), the neural and pigmented retinas (R) (b) are of the quail type. The superficial ectoderm (EC) (a) is of quail type. The periocular mesenchyme from which the choroid (Ch) and the sclera (SC) will later arise is chimeric (b). In (c) the cornea1 endothelium is chimeric. (a) Bar = 100 pm. (b) Bar = 10 pm. (c) Bar = 10 pm. (Quail nucleus, double arrows; chick nucleus, single arrow).

FIG. 4. (a) Transverse section of the embryo (E5.5) represented in Fig. 1 at the level of the telencephalon and eye vesicles (Bar = 1 mm). (b) Detail of the roof of the telencephalon (Tel). Note that the primordium of the leptomeninx is chimeric (quail nucleus: double arrows; chick nucleus: single arrow) (Bar = 10 pm). (c) Diencephalic floor epithelium with the anlage of the neurohypophysis (Nh). Note the chimerism of the meninges (Bar = 10 pm). (d) Epiphysis (ep) and diencephalic (Di) roof of quail type (Bar = 10 pm). (e) The limit between donor (right) and host (left) neural epithelium is located in the di-mesencephalon (Mes) transition area (Bar = 10 Mm). (f) Hypothalamus (Hy) and adenohypophysis (Ah) are of the quail donor type. The mesenchyme is of the chick host type (arrows) (Bar = 10 pm).

204

DEVELOPMENTALBIOLOGY TABLE 1

EXPERIMENTAL DESIGNS AND NUMBER OF EMBRYOS STUDIED FOR THE MAPPING OF THE EARLY NEURAL PRIMORDIUM

EXPERIMENTS

SACRIFICE

VOLUME120,1987

ment IV in the article by Couly and Le Douarin (1985) and will only be rapidly recalled here: the olfactory placode and nerves (including Schwann cells), part of the stomodeal roof, the epithelium of the nasal cavity and, in the brain, the floor of the telecephalon were of graft origin. In this case, no mesenchymal cells of quail type were ever found in the host. In experiment IIIB, the superficial ectoderm of the upper beak, the egg tooth, the epithelium lining the distal part of the nasal cavity (external nasal aperture and vestibular concha) plus the epithelium covering the rostral area of the premaxillary region originated from the graft (see experiment III2 and III3 in Couly and Le Douarin, 1985) (Fig. 8b). In addition, the laterodorsal region of the telencephalon was derived from the neural plate adjacent to zone B of the fold (Figs. 8c, d and 9). Experiment IIIC involved the fold and plate territory caudal to that of IIIB over a length of about 150 pm (experiment IIIB just described). An area of superficial ectoderm on the top of the head, the caudodorsal region of the cerebral hemisphere on the operated side, the roof of diencephalon (including the hemiepiphysis) were of graft origin. Some mesenchymal cells were also observed in the rostra1 region of the head but in no case were these mesenchymal structures entirely derived from the graft (Figs. lOa-d and lla-d). Experiment IV The median and rostra1 regions of the neural plate plus the adenohypophysis territory lying in the rostromedian neural fold were grafted as indicated in Table 1 (experiment IV) and in Fig. 12a. As expected, the adenohypophysis and hypothalamus were labeled as was the neurohypophysis; the territory of the latter can therefore be localized just caudal to that of the infundibulum (Figs. 12b-e). Morphogenetic

*Stage of the embryos (evaluated by the number of somites formed) when the operation was performed. **Experiment IIIA: see Couly and Le Douarin (1985).

of quail cells on the operated side. A general view of the graft derived mesenchymal and brain structures is represented in Fig. 7. Experiment III. In experiment III (Fig. 8a), the grafts were positioned in such a way as to determine the derivatives originating from specific rostrolateral areas over a length of 450 pm of the three-somite-stage neural primordium (Table 1). Experiment IIIA, consisting in the substitution of the rostrolateral neural plate and ridge territory over a length of 150 pm has already been described as experi-

Movements of the Neural Pm’mordium

Y A double graft involving the adenoExperiment hypophysis territory and that corresponding to region B of the fold with the adjacent neural plate (as in experiment IIIB) was performed on the same embryo (Fig. 13a). The purpose of this design was to visualize, at different stages of development, the relative positions of these territories. As shown in Fig. 13b, the presence of two implants did not disturb the morphogenetic movements of the neural primordium. Moreover, the infolding of the rostra1 region of the neural plate appeared clearly when the localization of the grafted tissues was determined at sequential developmental stages (Figs. 13c-k). origin

of the Prosencephalic

Meninges

Experiment VI. The embryos were at the four-somite stage and the grafted area corresponded to the fold cau-

COULY AND LE DOUARIN

Avian

Neural

Primordium

205

FIG. 6. Experiment II. (a) Scanning electron microscopic view of thl e rostra1 region of a three-somite stage chick embryo showing th e rej Jion replaced by its quail counterpart. (b) Facial view of an operated em1nyo sacrificed at E3.5. Note that the grafted area (right side) is sma rller than the chick one (left side). A symmetrization of the facial structr lres is established at later stages (Bar = 1 mm). (c) Lateral vie?w of the same embryo on the right side showing the perfect incorporation of 1the graft in the lateral head structures (Bar = 1 mm). (d) Sectio Nnin the third ventricle (III) of the chimera with hypothalamic epithelium of donor and host (Bar = 10 pm). Q, quail cells; C, chick cells.

da1 to zone C, i.e., roughly to the presumptive mesencephalic region. Two host embryos were recovered and sacrificed at E5.5 and E8.5 (Fig. 14a). A large number

of mesectodermal cells of graft origin were found in 1;he hemifacial area on the operated side. In the eye rBegil on, the choroid and sclera were derived entirely fro lm 1ihe

206

DEVELOPMENTAL BIOLOGY

.,:...:. .:..;. .:..: . . . ....‘.‘. ‘.. :.0:...’ Q

a

u

b

FIG. 7. Schematic representation of graft and host contribution to the head mesenchyme (a, b) and brain (c, d) in an embryo at E5.5. Compare with Fig. 2.

graft (Fig. 14b). The interstitial adenohypophyseal connective tissue was also of quail type (Fig. 14c), as were the presumptive leptomeninx and pachymeninx throughout the telencephalon and diencephalon (Fig. 14d). It is noteworthy that in the prosencephalic meninges the blood vessel walls, including the pericytes and the musculoconnective elements, were of graft origin (Fig. 14e). The endothelial cells lining the lumen were, in contrast, always of host type. So, too, were the meninges at the level of the mesencephalon itself. Mesenchymal cells of the nasofrontal bud which give rise to the upper beak and those of the frontal area were derived predominantly from the graft (Fig. 14f). No quail cells were found in the superficial ectoderm. A small area of the rostromedial mesencephalic roof was of graft origin at E5.5 (Fig. 14g). Figure 15 recapitulates the fate and positioning of the implant derived cells. CONCLUSIONS

AND

VOLUME 120, 1987

As can be seen in Fig. 16, the presumptive adenohypophyseal hypothalamic region is located mediorostrally in the neural primordium, just anteriorly to the region yielding the neurohypophysis. The latter is flanked by the neuroepithelial areas which later form the optic vesicles. The territories from which the telencephalon arises are bilaterally and rostrally located with respect to the above mentioned structures. They are lined externally by that part of the fold from which develop (1) the epithelium of the olfactory cavities including the sensory olfactory placodes (Zone A) (2) the vestibular epithelium of the nasal cavity, the epidermis of the nasofrontal area and the beak (including egg tooth) (Zone B). Caudally are located the anlagen of the thalamus (in the plate) and epiphysis (laterally), while the corresponding neural fold yields the epidermis covering the forebrain (Zone C). It is interesting to correlate the close topographical relationships between the prosencephalic neural primordium, the adenohypophysis, the olfactory organs, and the facial ectoderm with certain congenital pathologies in humans. A case in point in De Myers’s mediofacial syndrome (1967), in which malformations of the diencephalotelencephalic regions (also called holoprosencephalies) are associated with nasofrontopremaxil-

b

DISCUSSION

These experiments enable us to draw up a map of the presumptive forebrain territories on the avian neural plate at the early somitic stages (Fig. 16). In our previous article (Couly and Le Douarin, 1985) we presented information concerning the rostrolateral neural folds and adjacent neural plate regions. In the additional grafting experiments reported here we have extended our investigations to the prosencephalic plate and rostra1 mesencephalic neural fold.

FIG. 8. Experiment III. (a) Scanning electron micrograph showing the territories involved in the operation in experiments III A, B, and C (see Table 1). (b-d) Diagrammatic representation of the further localization of the grafted cells in the superficial ectoderm (b) and teleneephalon (c, d), in experiment IIIB.

FIG. 9. Illustration of experiment IIIB on a chimera at E8.5. (a) General view of a horizontal section in the telencephalon showing the chick area (C), a transition zone with chick and quail cells, (C, Q), and a quail area (Q). Framed regions are represented in detail in (b-d) (Bar = 1 mm). (b) The rostra1 telencephalic region is made up of chick cells (Bar = 10 pm). (c) Detail of the neural epithelium in the quail area (Bar = 10 pm). (d) Detail of the neural epithelium in the chick area close to the graft. The epithelial cells lining the ependymal canal are of chick host type. In contrast, the more extensive cells are chimeric, thus demonstrating a transversal migration of the cells in the neural epithelium (Bar = 10 pm). 207

DEVELOPMENTALBIOLOGY

VOLUME120, 1087

In experiment V, involving a double transplantation in the rostrolateral and lateral areas of the proencephalon, respectively, the extensive movements of the presumptive brain areas with respect to each other could be well perceived. It appeared that the laterorostral regions of the fold yielding the olfactory tractus rapidly reach a ventral position while the lateral regions of neural plate will become rostrally positioned to form the telencephalon. In the experiments where the caudal prosencephalic and the mesencephalic folds were grafted, the meninges covering the telencephalon and the mesectoderm of the nasofrontal region along with the sclera and choroid membrane of the eyes were labeled by the quail nuclear marker. These results are in agreement with those reported by Johnston (1966), Noden (1978), and Le Lievre (1976) (see also Le Douarin, 1982). The Sturge-Weber syndrome described in humans (Alexander, 1972) is characterized by abnormalities of dermal vascularization of the face resulting in “portc wine” stained areas covering (partly or totally) the skin FIG. 10. Experiment 111~:Diagrammatic representation of the graftregions innervated by the facial branch of nerve V (triderived tissue in the brain (a, b) and mesenchyme (which is chimeric) geminal). The dermal vascular abnormality is, in some (c, 4. cases, concommitant with homolateral telencephalic vascular anomalies of the leptomeninx and choroidal lary hypoplasia. Other examples include adenohypo- vascular lesions associated with glaucoma. According to Alexander (1972) and Enjolras et al. physeal deficiencies, revealed by an insufficient produc(1985), the risk that leptomeninx and ocular angiomas tion of growth hormone, associated with nasofrontal malformations (Couly et u,!.,1982) and De Morsier’s syn- might be associated with the facial dermis abnormalities drome (1968) in which anosmia is associated with a exists only when the latter affect the upper facial region (i.e., naso frontal), the dermis of which originates from functional genital deficiency of hypothalamohypophythe pro- and mesencephalic crest. When the port-wine seal origin. The total length of the presumptive prosencephalic stain is on the mandibulary or neck regions, occurrence region of the neural plate at the three-somite stage is of angiomas in the meninges and ocular lesions has never about 450 pm. The neural crest area, which extends vir- been described. It is striking that the telencephalic leptually along the whole neural axis, except the rostra1 tomeninx and choroid of the eye also originate from the prosencephalon, begins in the future diencephalic neural same level of the neural crest, while in the posterior fold (zone 12 of Fig. 16). This region of the crest, together brain and spinal cord meninges arise from the mesoderm with the mesencephalic fold, provides the frontal, nasal, (see Le Douarin, 1982). The naevus fusco-caeruleus ophtalmo-maxillaris deand maxillary facial regions with mesectodermal mesenchyme (forming bones and dermis) as already shown scribed by Ota (1939) is characterized by pigment dein our laboratory and others (Le Lievre and Le Douarin, position on the same territories as the port-wine maculae 1975; Le Libvre, 1974,1976,1978; Johnson et a& 1979 and of the Sturge-Weber syndrome. In some cases, the two abnormalities coexist (Yoshida, 1952). It is interesting reviewed in Le Douarin, 1982), through quail-chick to note that, when this occurs, two derivatives of the transplantations carried out at later developmental neural crest are affected, the mesectodermal mesenstages. Conformational changes in the neural primordium are thyme lining the blood vessels and the pigment cells. Moya-Moya, described by Nishimoto and Takeuchi dramatic during the late neurula stage when the tube closes and morphogenesis of the encephalic vesicles takes (1968), is a stenotic dysplasia of cerebral arteries. It has place. Studies at the cellular level by several authors been reported that it may be found in association with (Jacobson et al, 1979; Schoenwolf, 1985) have shown that typical symptoms of Von Recklinghausen neurofibrochanges in cell shape and relative positions rather than matosis (Tomsick et aZ.,1976), a neural crest disease. It is known that, in these regions of the head, the cell division and growth are responsible for these events.

COULY AND LE DOUARIN

Avian

209

Neural Primordium

9’

d

FIG. 11. Illustration of experiment 111~showing that the hemiepiphysis (Ep) alnd the hemiroof of the diencephalon (Di) are of the quail type (a, b). (a) Bar = 1 mm; (b) bar = 10 Frn. In (c) the transition between graft and ho!st neural epithelium is visible. Note the transversal migration of quail cells within the chick area (double arrows) (Bar = 10 pm). (d) The epide !rmis overlying the diencephalon is chimeric (Bar = 10 pm).

FIG. 12. Experiment IV: (a) Scanning electron microscopic representation of the territories involved in transplantation: i.e., the “hypothalamoadenoposthypophyseal” anlage. (b, d) Transverse sections in an E5.5 experimental embryo at the level of hypothalamus (b) and neurohypophysis (d) (Bar = 1 mm). (c, e) Details of hypothalamus (c) and neurohypophysis (e). Note in (e) that the caudal level of the hypothalamus is of the chick type (Bar = 10 pm). 210

COULY AND LE DOUARIN

Avian

Neural

i

P&nordium

i

FIG. 13. Experiment V: (a) Double transplants indicated in scanning electron micrographs. B: zone of Fig. 8. ANR: anterior neural ridge. (b) Experimental embryo at E5.5: note the normal gross anatomy of the head of this embryo carrying a double transplant. The grafted territories are indicated in the hatched areas (Bar = 1 pm). (c-f) Evolution of graft positions in a schematic embryo at 3- and 5-somite stages (c, d) and at the lo-somite stage in (e, rostra1 view) and (f, ventral view). (g-j) Positioning of graft-derived territories at E3.5 (g, facial grafted territory; h, brain grafted territory) and E5.5 (i, j, facial and brain grafted areas).

FIG. 14. Experiment VI: graft of the rostra1 mesencephalic neural ridge. (a) Sagittal section of an experimental embryo at E8.5 (Bar = 1 mm). (b) Labeling of the choroid (Ch) and the sclera (SC) by quail cells (Bar = 10 pm). (c) Adenohypophysis glandular cords (Ah) with quail mesenchymal cells (double arrows) (Bar = 10 pm). (d) Leptomeninx (Lm) of the telencephalon (Tel) is made up of quail cells (double arrows) (Bar = 10 pm). (e) The connective wall (but not the endothelial cells) of the blood vessels of meninges are of the quail type (double arrows) (Bar = 10 pm). (f) The mesenchyme of the nasal bud is made up of quail cells (Bar = 10 pm). (g) A small longitudinal area in the dorsomedian tectum opticum (To) is made up of quail cells (Bar = 10 pm). 212

COULY AND LE DOUARIN

a

b

FIG. 15. Schematic representation of the mesectodermic area labeled in experiment IV. (a) Migration of mesectodermal cells of quail into the nasofrontal and periocular regions. (b) The meninges of the prosencephalon are mesectodermal origin and derive from the caudal prosencephalic and rostra1 mesencephalic neural crest.

vascular endothelium does not originate from the mesectodermal mesenchyme but from mesodermal cells belonging to the hemangioblastic cell lineage (Le Lievre and Le Douarin, 1975,PQault et al, 1983). Vascularization of these territories requires their invasion by endothelial cells originating from the branchial arch arteries ( our unpublished results). It is very likely, although not yet demonstrated, that such a process requires the production, by the mesectodermal mesenchyme, of angiogenetic factors responsible for attraction and growth of the vas-

Avian Neural Priwdium

213

cular endothelial cells. Production of such factors must be controlled during organogenesis and maintained at a certain level to regulate the extent of angiogenesis. We propose as a hypothesis that in the angiomas a defect in this regulation leads to hyperproduction of the angiogenetic factor. A somatic mutation could have occurred in one of the early precursors of the pro- and mesencephalic mesectodermal cells whose clonal progeny subsequently extend to both the meningeal (including the eye choroid) and facial dermis regions. When such an event occurs in the neural crest cells of the hindbrain neural fold, the abnormality is restricted to the dermis and does not therefore present the dramatic evolution resulting from the meninges defect since no meninges arise from this region of the neural crest.

The authors thank Monique Le Thierry for her skilful technical assistance, Bernard Henri and Sophie Tissot for their help in the iconography, and Evelyne Bourson for typing the manuscript, This work was supported by the Centre National de la Recherche Scientifique and by grants from the Institut National de la Sante de la Recherche Medicale, the Minis&e de la Recherche et de l’Industrie, the Fondation pour la Recherche Medicale Francaise, and the Ligue Francaise contre le Cancer and by a Basic Research Grant 1-866 from March of Dimes Birth Defects Foundation. Dr. G. F. Couly is a maxillofacial pediatric surgeon at Hopital Necker-Enfants Malades in Paris.

FIG. 16. Mapping of the anterior neural primordium at the three- to four-somite stage in the avian embryo (Bar = 100 pm). (1) Adenohypophysis, (2) hypothalamus, (3) ectoderm of nasal cavity, (4) floor of telencephalon, (5) olfactive placode, (6) ectoderm of upper beak and egg tooth, (7) optic vesicles, (8) neurohypophysis, (9) roof of telencephalon, (10) diencephalon, (11) hemiepiphysis, (12) ectoderm of calvaria and caudal prosencephalic neural crest (light spotted area), (13) rostra1 mesencephalic neural crest (dense spotted area), (14) mesencephalon.

214

DEVELOPMENTALBIOLOGY REFERENCES

ALEXANDER,G. L. (1972). Sturge-Weber syndrome. Hand Cl& Neural 14,223-240.

COULYG. F., and LE DOUARIN,N. M. (1985). Mapping of the neural early primordium in quail-chick chimaeras. I. Developmental relationships between placodes, facial ectoderm and prosencephalon. Dev. Biol. 110.422-439. COULY,G., RAPPAPORT,R., BRAUNER,R., and RAULT, G. (1982). Association of facial malformations with primary hypopituitarism. Pediatr. Res. 182,886-906. DE MORSIER,G. (1967).Agenesie des lobes olfactifs et des commissures calleuses et antkrieures. La dysplasie olfactogdnitale. In “Etude sur les malformations du cerveau” (G. De Morsier, Ed.), pp. 101-153. MQdecine et Hygiene de Geneve. DE MYER, I. (1967). The median cleft face syndrome. Neurology 17, 961-972. ENJOLRAS,O., RICHE, M. C., and MERLAND,J. J. (1985). Facial portwine stains and Sturge-Weber syndrome. Pediatrics 76,48-51. HAMBURGER,V., and HAMILTON,H. L. (1951). A series of normal stages in the development of the chick embryo. J. Morphol. 88,49-92., JACOBSON,A. G., MIYAMOTO,D. M., and MAI, S. H. (1979). Rathke’s pouch morphogenesis in the chick embryo. J. Exp. Zool. 207, 351366. JOHNSTON,M. C. (1966). A radioautographic study of the migration and fate of cranial neural crest cells in the chick embryo. Anat. Rec. 156.143-156.

JOHNSTON,M. C., NODEN,D. M., HAZELTON,R. D., COULOMBE, J. L., and COULOMBE, A. J. (1979).Origins of avian ocular and periocular tissues. Exp. Eye Res. 29,27-43. LE DOUARIN,N. M. (1982). “The Neural Crest.” Cambridge Univ. Press, Cambridge.

VOLUME120, 1987 LE LI~~VRE,C. (1974). Role des cellules mesectodermiques issues des cretes neurales cephaliques dans la formation des arcs branchiaux et du squelette visceral. J. Emlnyol. Exp. Morphol. 31,453-477. LE LII?VRE,C. (1976). “Contribution des c&es neurales a la genese des structures cbphaliques et cervicales chez les oiseaux.” These d’Etat, Nantes. LE LI~VRE, C. S. (1978). Participation of neural crest-derived cells in the genesis of the skull in birds. J. Embryol. Exp. Morph01 47,1737. LE LIBVRE,C., and LE DOUARIN,N. M. (1975).Mesenchymal derivatives of the neural crest: Analysis of chimaeric quail and chick embryos. J. EmbyoC Exp. Mwphol 34,125-154. NISHIMOTO,A., and TAKEUCHI,S. (1968). Abnormal cerebral network related to the internal carotid arteries. J. Neural 29,255-260. NODEN,D. M. (1978). The control of avian cephalic neural crest cytodifferentiation. I. Skeletal and connective tissues. Dev. BioL 67,296312. OTA, (1939). Naevus fusco-caeruleus ophtalmo-maxillaris. Tokyo Med J. 62,1243-1245. P~AULT, B. M., THIERY, J. P., and LE DOUARIN,N. M. (1983). Surface marker for hemopoietic and endothelial cell lineages in quail that is defined by a monoclonal antibody. Proc. Nat1 Acad. Sci USA 80, 2976-2980. SCHOENWOLF, G. C. (1985). Shaping and bending of the avian neuroepithelium: Morphometric analyses. Dew. Biol. 109,127-139. TOMSICK, T. A., LUKIN, R. R., CHAMBERS, A. A., and BENTON,C. (1976). Neurofibromatosis and intracranial arterial occlusive disease. Neurwradiologg 11,229-234. YOSHIDA, N. (1952). Naevus fusco-caeruleus ophtalmo maxillaris. Tohoku J. Exp. Med. 55, (Suppl.), l-104. ZACCHEI, A. M. (1961). Lo sviluppo embrionale della quaglia giapponese (Coturnix oturnix japonica T. e S.) Arch. Ital. Anat. Embriol. 66, 36-62.