Using ‘yolk sac chimaeras’, we have previously demonstrated that stem cells, destined to colonize haemopoietic organs other than the yolk sac, arise in the embryo proper. We have now investigated the emergence and potentialities of these cells in vivo and in vitro.

The in vivo approach consisted of interspecies grafting between quail and chick embryos. The cell progeny from the grafts was detected by means of QH1, a monoclonal antibody specific for the quail haemangioblastic lineage. When grafted into the dorsal mesentery of the chick embryo, which is a haemopoietic microenvironment, the region of the aorta from E3-E4 quail embryos generated large haemopoietic foci. When associated with a chick attractive thymic rudiment, cells left the quail aorta, entered this rudiment and underwent lymphopoiesis.

Cell suspensions prepared from 40-50 chick aortae, seeded in appropriate semi-solid media, yielded macrophage, granulocyte or erythrocyte clones. These colony forming cells were two to eight times more frequent than in cell preparations from hatchling bone marrow. By contrast, cells prepared from the whole embryonic body deprived of the aorta were not clonogenic.

By interspecies grafting of somatopleural (ectoderm + mesoderm, e.g. limb bud) or splanchnopleural rudiments (endoderm + mesoderm, e.g. lung, pancreas, intestine), the endothelial lining of blood vessels was shown to arise by two entirely different processes according to the rudiment considered: angiogenesis, i.e. invasion by extrinsic endothelial cells, in the limb bud, and vasculogenesis, i.e. in situ emergence of endothelial cells, in internal organs. The spleen, which first develops as a continuum to the pancreatic mesoderm, acquires its endothelial network by vasculogenesis, and is colonized by extrinsic haemopoietic stem cells. Granulopoietic cells in the pancreas and accessory cells in the lung are also extrinsic. Thus, in the case of endomesodermal rudiments, interspecies grafting reveals separate origins of endothelial and haemopoietic cells.

The emergence during embryonic life of the stem cells that colonize the rudiments of haemopoietic organs and found definitive blood cell lineages is still enigmatic in many respects. The process first occurs in the chick yolk sac during day 2 (E2) of incubation. At that time, solid groups of cells become committed to the ‘haemangioblastic’ lineage, which comprises haemopoietic and endothelial cells and their precursors. Within a few hours of their appearance, the central cells in these groups detach and evolve towards the erythroid pathway, while the peripheral cells take on the aspect of endothelium (Sabin, 1920; Kessel & Fabian, 1985; Pardanaud et al. 1987a; Péault et al. 1988). A new process initiated towards the end of day 2 (Dardick & Setterfield, 1978) culminates in the release of a different generation of red cells on day 5, at which time haemopoiesis becomes active in the embryo proper, beginning with the first colonization period of the thymic rudiment (reviewed by Le Douarin et al. 1984).

Using a chimaera model, consisting of a quail embryo grafted onto a chick yolk sac, we have demonstrated that this second phase of haemopoiesis depends on the emergence of a new set of stem cells, which are produced in the embryo proper rather than in the yolk sac (reviewed by Dieterlen-Lièvre, 1984).

We are now addressing the following questions. (1) What are the site of origin and potentialities of the intraembryonic stem cells that relay yolk sac stem cells? (2) Is there a common precursor for haemopoietic stem cells and endothelial cells?

The existence of this precursor, the haemangioblast according to the term coined by Murray (1932), has been postulated on descriptive grounds. First, a frequent observation in 1 to 3-day embryos, both in yolk sac blood islands and in intraembryonic blood vessels, is round cells that appear to have budded off from the endothelium (Sabin, 1920; Dieterlen-Lièvre & Martin, 1981; Péault et al. 1988). Second, endothelia and haemopoietic cells share common antigens, recognized in the quail by monoclonal antibodies (mAb). These antigens are strongly immunogenic, so that immunization with different preparations has induced the synthesis of several mAbs with a similar specificity for the quail hemangioblastic lineage. Two of these mAbs have been selected and extensively used in experiments with avian chimaeras. MB I (Péault et al. 1983; Péault, 1987) was obtained from immunization with the quail immunoglobulin µ chain. MB1 recognizes several glycoproteins of apparent molecular masses ranging from 80 to 2OO×103, among which is a2 macroglobulin. The second antibody of this type, QH1 (Pardanaud et al. 1987a), has been obtained following immunization with 12-day embryonic quail bone marrow. Like MB1, it recognizes the haemangioblastic lineage of the quail and no cells of the chick. In Western blots carried out with bone marrow cell extracts, it recognizes several molecules ranging inM1. between 150 and 220 (× 103) (Pardanaud et al. 1987α). Both antibodies have similar specificities, though QH1 also recognizes quail primordial germ cells (Pardanaud et al. 1987ò).

In the present report, we will consider two lines of experimental data; the first results from interspecific grafting of rudiments between quail and chick, followed by tracing cells of quail origin with antibody QH1. By this means: (1) stem cells that could be transplanted, and which displayed various haemopoietic potentialities, were shown to arise at E3-E4 in the para-aortic region; (2) evidence was obtained that some rudiments acquire independently their contingents of endothelial and haemopoietic precursors.

The other class of results reviewed here relies on the analysis of haemopoietic potentialities present in various regions of the early embryo by the clonal culture system. This system has yielded definite proof of the primordial role of the paraaortic region in the production of intraembryonic stem cells (Cormier et al. 1986; Cormier & Dieterlen-Lièvre, 1988).

Haemopoietic stem cells in the E3-E4 para-aortic region

Diffuse haemopoiesis occurs in the dorsal mesentery at E5-E7 in the quail embryo and at E6-E8 in the chick embryo (Miller, 1913; Dieterlen-Lièvre & Martin, 1981). We thought that this site might be a favourable microenvironment for the multiplication and differentiation of cells with haemopoietic potentialities and we devised a technique for grafting rudiments to this site.

During normal development, the process of cell budding observed in various vessels is maximal in the dorsal aorta at E3. Dense mats of basophilic cells can be seen aggregated to the ventral aspect of the endothelium towards the lumen. This process also occurs on the tissue side of the aortic endothelium. Furthermore, similar basophilic cells (QH1+ in the quail) are also scattered within the meshes of the dorsal mesenteric mesenchyme. In yolk sac chimaeras these cells are quail (DieterlenLièvre & Martin, 1981). The aorta thus appears to be a likely candidate capable of producing the diffuse haemopoietic foci present in the paraaortic region.

The ‘thoracic’ segment of the dorsal aorta in the avian embryo is a large vessel, uncluttered by neighbouring organs between the levels of the aortic arches and the pro-/mesonephros. Until E5, the aortic endothelium in this free portion is surrounded by a muff of undifferentiated mesodermal cells. The whole segment with its mesodermal envelope can be dissected away with relative ease at this stage. When inserted ventral to the chick host aorta, the graft integrates into the mesentery. Three days later, it has given rise to a dense aggregate of cells that display the quail nucleolar marker and are positive with the quail haemangioblastic marker QH1 (Fig. 1). Exceptionally, a structure reminiscent of the engrafted vessel can be observed budding positive cells in the lumen (Fig. 2). Endothelial cells are sometimes present and these participate in the formation of chimaeric endothelial structures. The aggregates of grafted haemopoietic cells are located in the vicinity of host blood-forming foci. The cells in these endogenous foci undergo erythropoiesis and less frequently granulopoiesis. Grafted quail cells usually differentiate as erythroblasts.

To determine whether progenitors with other potentialities were present in this region, the quail aorta, as defined above, was associated with a chick thymic rudiment. The thymus was taken at E6.5, when it becomes receptive to the first wave of colonizing stem cells. The two rudiments were co-cultured on an agar nutritive medium (Wolff & Haffen, 1952) for 12 h prior to grafting for 9 days in a chick host. Sequential sections of the transplants were treated with QH1 mAb and CT1 mAb, which is specific for thymocytes (Chen et al. 1984). Out of 10 associations, all but one harboured QH1+ cells. Some of the thymic lobes that developed in each transplant were entirely populated by these quail cells (Fig. 3), while others were chimaeric. The CT1 antibody marked all thymocytes, irrespective of whether they were derived from the chick host or the quail aorta (Fig. 4). This experiment clearly demonstrates that precursors with T lymphocyte potentialities are present in the para-aortic region.

The question is still open as to whether haemopoietic cells deriving from the aortic rudiment are the progeny of either the endothelium or the mesoderm surrounding it. Only a partial answer can be given, because it is extremely difficult to dissect the endothelium cleanly. When the more or less bared endothelium is grafted, haemopoietic foci may develop, but their size is reduced.

Other rudiments were grafted to the same site: somites (from E2 embryos), liver, kidney, and another blood vessel, the common cardinal vein, (from E3 or E4). In none of these cases, did haemopoietic cells ever develop. The cardinal vein gave rise to a thin sheet of quail cells inserted in the mesentery of the chick host. The vessel wall did not disintegrate as in the case of the aorta; however, no lumen persisted in these non-functional vessels.

Haemopoietic stem cells in the E2 splanchnopleural mesoderm

In the E2 embryo, it is possible by mechanical means to dissect away the mesoderm either from the ectoderm or the endoderm with which it is associated. The splanchnopleural mesoderm from the area pellucida was divided into a number of segments along the cephalocaudal axis. When these segments were grafted to the chick dorsal mesentery, they gave rise to QH1+ cells with extensive migratory behaviour. Cells from the grafted aorta gave rise to a solid aggregate of differentiating blood cells at the site of grafting in the mesentery; whereas, by contrast, the progeny of splanchnopleural mesoderm spread to different organs or tissues, mainly the mesentery, liver and kidney. These progeny were modest in number and the cells, identified by virtue of their QH1 affinity, adopted diverse phenotypes, apparently dictated by the environment in which they became arrested. For instance, cells that lodged in the liver or the kidney participated in the formation of chimaeric endothelia; in each location they became incorporated into the specific type of endothelium that characterizes that organ (Figs 5—6). Such specification imposed on invading foreign endothelial cells has been observed in grafts of mouse brain to the chick chorial1antoic membrane (Risau et al. 1986). Cells that settled in the dorsal mesentery retained the round shape and the isolated status of haemopoietic cells (Dieterlen-Lièvre, 1984).

In some experimental series, mesodermal cells were not dissociated from the endoderm, so that the grafts contained cells from the two germ layers. Their fate was entirely different from that of isolated mesoderm. These grafts became implanted in the coelom and usually formed intestinal-like structures, associated with haemopoietic cell groups. These cell groups were dense aggregates of QH1+ cells, reminiscent of aorta-derived haemopoietic cells. The splanchnopleural area corresponding to the level of somites 15-25 (Fig. 7) gave a particularly abundant QH1+ progeny. This level is precisely the one that, in various types of interspecies chimaeras involving the in ovo transplantation of blastodisc territories, was found to be important for the formation of haemopoietic foci in the mesentery (Martin et al. 1980). In some of the splanchnopleural explants, the QH1+ aggregates were spheres tightly bound by an endothelial envelope, i.e. blood islands.

Comparison of the derivatives yielded by the E3-E4 aortic region, E2 mesoderm or E2 mesoderm + endoderm indicates a positive influence of endoderm on the proliferation and, perhaps, the commitment of mesodermal cells to the haemopoietic pathway. Such an influence has been demonstrated earlier by Miura & Wilt (1970) in the case of extraembryonic endoderm, which promotes in vitro the growth of blood islands in the area vasculosa mesoderm.

Assessment in clonal cultures of the concentration and types of progenitors in the para-aortic region

The art of cloning avian blood progenitors has lagged behind that achieved for mammalian cells. In particular, most growth factors remain unidentified. Mammalian growth factors do not cross-react functionally on avian haemopoietic cells. Usual media contain chicken serum, which favours powerfully the growth of macrophages and may even commit multipotential precursors to the monocytic pathway (Dodge & Sharma, 1985; Cormier & Dieterlen-Lièvre, 1988). The only avian growth factor available, c-MGF (Leutz et al. 1984), acts on monocytic precursors. Chicken anaemic serum is the usual source of avian erythropoietin. Our main goal was to detect whether progenitors with various potentialities were present in the wall of the aorta. To this end, Cormier devised a medium with only a minimal amount of serum that permits the growth and differentiation of granulocytic clusters. Another medium, supplemented with fibroblast-conditioned medium, supports the growth of monocytic and granulomonocytic colonies.

With the culture conditions available, we have been able to show that progenitors for monocytic, granulocytic and erythroid colonies are present in the wall of the aorta at E3 and E4 (Cormier et al. 1986; Cormier & Dieterlen-Lièvre, 1988). Strikingly, cell suspensions prepared from the embryo deprived of its aortic region yielded no progenitors. Furthermore, the progenitor concentration is two to eight times higher in the embryonic aorta wall than in the hatchling bone marrow (Fig. 8). We have not determined yet whether, depending on the culture conditions, the same precursors differentiate along various pathways or whether different, committed, precursors are recruited. In the latter case, the number of progenitors in the aortic wall would amount to the sum of individual precursors detected. We are devising culture conditions to search for pluripotential progenitors. This should enable us to enumerate more precisely the haemogenic capacities of this region. In any event, the clonal culture system yields critical evidence for the primordial role of this region in the emergence of haemopoietic stem cells at the period when yolk sac stem cells are first relayed by intraembryonic cells.

Vasculogenesis, angiogenesis and haemopoiesis

Following Risau et al. (1988), it is convenient to distinguish vasculogenesis, the process by which endothelial cells arise de novo, from angiogenesis, the process by which normal or tumour tissues are invaded by pre-existing endothelial cells whose proliferation is stimulated by these tissues. We are currently mapping these processes in the developing embryo and attempting to relate them to the process of haemopoiesis.

The use of QH1 antibody on whole mounts (Fig. 7) (Pardanaud et al. 1987α) and of MB1 on sections (Péault et al. 1988) has revealed the very first steps of vasculogenesis, which were not distinguishable without a specific marker of endothelial cells. The two studies contribute different information on the emergence of the first endothelial cells. Vasculogenesis is initiated in the blastodisc at the head process stage (Pardanaud et al. 1987a). The first positive cells appear in the area opaca at a median level, lateral to the first pairs of somites, and about 2h later in the area pellicuda at the same level. Cells, first isolated, subsequently link up to form an extending network, which invests the blastodisc simultaneously in the cephalic and caudal directions. Caudalwards vascularization progresses parallel to segmentation. The caudal ‘horseshoe’ area vasculosa becomes established at the stage with five pairs of somites.

QH1+ cells appear to emerge in situ all over the blastodisc; experimental evidence that there is no privileged site of formation of endothelial cells has also been obtained from the analysis of yolk sac chimaeras, in which quail and chick territories were associated surgically (Dieterlen-Lievre, 1975; Martin et al. 1978). In these chimaeras the endothelia did not migrate from one area to another: in the chick yolk sac territory, endothelia were chick, while in the quail embryo territory endothelia were quail. Thus it is clear that no extensive migration of endothelial cells occurs in the horizontal plane of the blastodisc.

However, the vertical investiment of the germ layers by endothelial cells proceeds differently. The pictures published by Péault et al. (1988) show that isolated MB1 + cells first appear in ventral apposition to the mesoderm in contact with the endoderm.

By contrast, the somatopleural layer of the mesoderm is devoid of positive cells both in the area vasculosa and in the area pellucida. Péault et al. described the differentiation of intersegmental arteries by a proliferation of the aortae. These adventitious vessels penetrate between the somites, surround and eventually penetrate the neural tube. Whether endothelial cells also emerge in situ from the somatopleural mesenchyme cannot be surmised from these descriptive data.

In this regard it is interesting to note that the amnion, an ecto-mesodermal derivative, never acquires blood vessels. These considerations led us to study the respective origins of endothelial and haemopoietic cells in rudiments consisting of either endoderm or ectoderm associated with mesoderm. For the ecto-/mesodermal rudiment, the limb bud was selected. Jotereau & Le Douarin (1978) had previously demonstrated that haemopoietic cells, endothelial cells and osteosclasts were of host origin in chick limb buds explanted at stages 23-25 (Hamburger & Hamilton, 1951) and grafted to the somatopleura of quail hosts. To determine the host or graft origin of cells, these investigators used the quail nucleolar marker, which is difficult to identify in endothelial cells; furthermore, isolated cells from one or the other species cannot usually be detected. These problems are avoided by the use of monoclonal antibodies. Chick limb buds grafted to quail hosts displayed a rich QH1+ endothelial network around cartilaginous condensations (Fig. 9). Bone marrow harboured endothelial and haemopoietic cells of the quail (Fig. 10). In the reverse association (quail limb buds grafted to chick hosts) any QH1+ cell must have entered the rudiment prior to its retrieval from the donor embryo. Quail limb buds were taken from donors between the stages with 27 and 43 pairs of somites (HH16-21). They were grown until a total age of 13-16 days, when bone marrow differentiation is well under way. Two features were noted: (1) the bone marrow never contained positive cells; (2) scarce positive cells could be observed in the connective tissue surrounding the cartilage masses; they were usually isolated and occasionally inserted in a chimaeric vessels. The number of these cells was always very low (around 10) and was not significantly related to the stage at which the rudiment was obtained. These experiments enable us to conclude that endothelial cells and haemopoietic cells in the bone marrow both have an extrinsic origin, and that their ingress occurs after E4.

Various endo-/mesodermal rudiments were also grafted (Table 1). Some of these organs are not considered to have a haemopoietic function, but they harbour blood cells that exert specific functions, for instance macrophages in the lung. As for the chick pancreas, it is a very active site for granulopoiesis during embryonic life (Benazzi-Lentati, 1932; Dieterlen-Lièvre, 1965). Finally, the rudiment of the spleen carries out erythropoiesis and granulopoiesis during embryogenesis. Though purely mesodermal, it arises as an appendix to the pancreas and is thus subjected to interactions with endoderm.

All these organs developed according to the same rule. Endothelial cells and haemopoietic cells were distinct in origin; the first arose from precursors intrinsic to the rudiment, while the latter, provided by the host, invaded the rudiment. Concretely, when the rudiments were of the chick species, endothelial cells were QHl‐ and haemopoietic cells were positive (Figs 11, 13, 15). When rudiments came from the quail, endothelial cells were QH1+ and haemopoietic cells were negative (Figs 12, 14, 16). The rudiments were obtained from donor embryos between 55 and 84 h of incubation, covering a range of stages between 25 and 43 pairs of somites (HH15 to -21). They were left to develop until they had reached a total age of 8-17 days. Regardless of the precise period of engraftment, the results were similar. It should be mentioned that chick rudiments grafted in quail hosts occasionally displayed a few endothelial profiles at the periphery of the explant (Fig. 11). This marginal ingression of host endothelia does not detract from the conclusion that the bulk of endothelial cell precursors arise within the mesoderm of these organs.

To summarize, interspecies grafting makes it possible to detect two independent processes during avian development: vasculogenesis in rudiments in which mesoderm is associated with endoderm, and angiogenesis in rudiments in which mesoderm is associated with ectoderm. Furthermore, haemopoiesis develops independently of vasculogenesis in the former rudiments.

We have dissected various events in the development of the avian haemopoietic system. First, a region of splanchnopleural mesoderm could be shown to produce haemopoietic cells. This region, though not precisely mapped yet, is located in the area pellucida, at the level of somites 10-25. Interactions between endoderm and mesoderm appear to exert a positive influence on the blood-forming properties of this region. It develops within 24 h in a tissue, surrounding the aorta, in which clonogenic cells of various blood lineages could be enumerated by culture in semi-solid media. Second, organ rudiments acquire their endothelial cell complement by two different processes, vasculogenesis or angiogenesis. Vasculogenesis, i.e. in situ emergence of endothelial cells, also appears to depend on the presence of endoderm interacting with the mesoderm. Where it occurs, haemopoietic stem cells have an independent origin from endothelial cells. Angiogenesis, i.e. colonization of a rudiment by extrinsic endothelial cell progenitors, occurs in the bone marrow, which is the definitive blood stem-cell reserve.

No experimental evidence is available concerning the link between endothelial and haemopoietic cell origins in the bone marrow. One of the questions about the development of the haemopoietic system deals with the existence of the haemangioblast, a putative common precursor to endothelial and haemopoietic lineages. If this precursor exists, its progeny diverge subsequently, as clearly indicated by the independent origins of the two lineages in endomesodermal organs.

We are indebted to Chen-lo Chen and Max Cooper for the kind gift of CT1 mab. This work was supported by the Centre National de la Recherche Scientifique and the Ministère de la Recherche et de la Technologie.

Benazzi-Lentati
,
G.
(
1932
).
Sul focolaio eosinofilopoietico del pancreas di Uccelli
.
Monitore Zool. Itai
.
43
,
115
118
.
Chen
,
C. H.
,
Chanh
,
T. C.
&
Cooper
,
M. D.
(
1984
).
Chicken thymocyte-specific antigen identified by monoclonal antibodies: ontogeny, tissue distribution and biochemical characterization biochemical characterization
.
Eur. J. Immun
.
14
,
385
391
.
Cormier
,
F.
,
De Paz
,
P.
&
Dieterlen-Lievre
,
F.
(
1986
).
In vitro detection of cells with monocytic potentiality in the wall of the chick embryo aorta
.
Devi Biol
.
118
,
167
175
.
Cormier
,
F.
&
Dieterlen-Lievre
,
F.
(
1988
).
The wall of the chick embryo aorta harbors M-CFC, G-CFC, GM-CFC and BFU-E
.
Development
102
.,
279
285
.
Dardick
,
I.
&
Setterfield
,
G.
(
1978
).
Early origins of definitive erythroid cells in the chick embryo
.
Tissue & Cell
10
,
335
364
.
Dieterlen-Lievre
,
F.
(
1965
).
Etude morphologique et experimentale de la differenciation du pancreas chez 1’embryon de Poulet
.
Bull. Biol. Fr. Belg
.
99
,
4
116
.
Dieterlen-Lievre
,
F.
(
1975
).
On the origin of haemopoietic stem cells in the avian embryo: an experimental approach. J
.
Embryol. exp. Morph
.
33
,
607
619
.
Dieterlen-Lievre
,
F.
(
1984
).
Blood in Chimeras
.
In Chimeras in Developmental Biology
(ed.
N.
Le Douarin
&
A.
McLaren
), pp.
133
163
.
London
:
Academic Press
.
Dieterlen-Lievre
,
F.
&
Martin
,
C.
(
1981
).
Diffuse intraembryonic hemopoiesis in normal and chimeric avian development
.
Devi Biol
.
88
,
180
191
.
Dodge
,
W. H.
&
Sharma
,
S.
(
1985
).
Serum-free conditions for the growth of avian granulocyte and monocyte clones and primary leukemic cells induced by AMV, and the apparent conversion of granulocytic progenitors into monocytic cells by a factor in chicken serum
.
J. cell. Physiol
.
123
,
264
268
.
Hamburger
,
V.
&
Hamilton
,
H. L.
(
1951
).
A series of normal stages in the development of the chick embryo
.
J. Morph
.
88
,
49
92
.
Jotereau
,
F. V.
&
Le Douarin
,
N. M
, (
1978
).
The developmental relationship between osteocytes and osteoclasts: A study using the quail-chick nuclear marker in endochondral ossification
.
Devi Biol
.
63
,
253
265
.
Kessel
,
J.
&
Fabian
,
B. C.
(
1985
).
Graded morphogenetic patterns during the development of the extraembryonic blood system and coelom of the chick blastoderm: a scanning electron microscope and light microscope study
.
Am. J. Anat
.
173
,
99
112
.
Le Douarin
,
N.
,
Dieterlen-Lievre
,
F.
&
Oliver
,
P.
(
1984
).
Ontogeny of avian primary lymphoid organs and lymphoid stem cells
.
Am. J. Anat
.
170
,
261
299
.
Leutz
,
A.
,
Beug
,
H.
&
Graf
.
T.
(
1984
).
Purification and characterization of cMGF, a novel chicken myelomonocytic growth factor
.
EMBO J
.
3
,
3191
3197
.
Martin
,
C.
,
Beaupain
,
D.
&
Dieterlen-Lievre
,
F.
(
1978
).
Developmental relationships between vitelline and intra-embryonic haemopoiesis studied in avian “yolk sac chimaeras”
.
Cell Differ
.
7
,
115
130
.
Martin
,
C.
,
Beaupain
,
D.
&
Dieterlen-Lievre
,
F.
(
1980
).
A study of the development of the hemopoietic system using quail-chick chimeras obtained by blastoderm recombination
.
Devi Biol
.
75
,
303
314
.
Miller
,
A. M.
(
1913
).
Histogenesis and morphogenesis of the thoracic duct in the chick; development of blood cells and their passage to the blood stream via the thoracic duct
.
Am. J. Anat
.
15
,
131
198
.
Miura
,
Y.
&
Wilt
,
F. H.
(
1970
).
The formations of blood islands in dissociated-reaggregated chick embryo yolk sac cells
.
Expl Cell Res
.
59
,
217
226
.
Murray
,
P.
(
1932
).
The development in vitro of the blood of the early chick embryo
.
Proc. R. Soc. Lond
.
111
,
497
521
.
Pardanaud
,
L
,
Altmann
,
C.
,
Kitos
,
P.
,
Dieterlen-Lievre
,
F.
&
Buck
,
C.
(
1987a
).
Vasculogenesis in the early quail blastodisc as studied with a monoclonal antibody recognizing endothelial cells
.
Development
100
,
339
349
.
Pardanaud
,
L.
,
Buck
,
C.
&
Dieterlen-Lievre
,
F.
(
1987b
).
Early germ cell segregation and distribution in the quail blastodisc
.
Cell Differ
.
22
,
47
60
.
Péault
,
B. M.
(
1987
).
MB1, a quail leukocyte/vascular endothelium antigen: characterization of the lymphocyte-surface form and identification of its secreted counterpart as a α-2-macroglobulin
.
Cell Differ
.
21
,
175
187
.
Péault
,
B.
,
Coltey
,
M.
&
Le Douarin
,
N. M.
(
1988
).
Ontogen1c emergence of a quail leukocyte/endothelium cell surface antigen
.
Cell Differ
.
23
,
165
174
.
Péault
,
B. M.
,
Thiery
,
J. P.
&
Le Douarin
,
N. M.
(
1983
).
Surface marker for hemopoietic and endothelial cell lineages in quail that is defined by a monoclonal antibody
.
Proc. natn. Acad. Sci. U.S.A
.
80
,
2976
2980
.
Risau
,
W.
,
Hallmann
,
R.
,
Albrecht
,
U.
&
Henke-Fahle
,
S.
(
1986
).
Brain induces the expression of an early cell surface marker for blood-brain barrier-specific endothelium
.
EMBO J
.
5
,
3179
3183
.
Risau
,
W.
,
Sariola
,
H.
,
Zerwes
,
H.-G.
,
Sasse
,
J.
,
Ekblom
,
P.
,
Kemler
,
R.
&
Doetschman
,
T.
(
1988
).
Vasculogenesis and angiogenesis in embryonic-stem-cell-derived embryoid bodies
.
Development
102
,
471
478
.
Sabin
,
F.
(
1920
).
Studies on the origin of the blood-vessels and of red-blood corpuscles as seen in the living blastoderm of chicks during the second day of incubation
.
Contributions to Embryology
, vol.
9
, pp.
214
262
.
Washington
:
Carnegie Instn
.
Wolff
,
E.
&
Haffen
,
K.
(
1952
).
Sur le développement et la différenciation sexuelle des gonades embryonnaires d’Oiseau en culture in vitro
.
J. exp. Zool
.
119
,
381
404
.