How studies on the avian embryo have opened new avenues in the understanding of development: a view about the neural and hematopoietic systems.

The chick embryo is as ancient a source of knowledge on animal development as the very beginning of embryology. Already, at the time of Caspar Friedrich Wolff, contemplating the strikingly beautiful scenario of the germ deploying on the yellow background of the yolk inspired and supported the tenants of epigenesis at the expense of the preformation theory. In this article, we shall mention some of the many problems of developmental biology that were successfully clarified by research on chick embryos. Two topics, the development of the neural system and that of blood and blood vessels, familiar to the authors, will be discussed in more detail.

The quest for hematopoietic stem cells in the embryo. An interview with Françoise Dieterlen-Lièvre. Interview by Thierry Jaffredo and Charles Durand.

Françoise Dieterlen-Lièvre is probably the scientist who has most contributed to our basic knowledge of developmental hematopoiesis. She has dedicated her career to answering cutting edge questions on the origin of hematopoietic stem cells in the embryo. Her seminal contributions, widely recognized by the scientific community, have paved the way for generations of developmental hematologists questioning the origins of hematopoietic stem cells. After having demonstrated the intra-embryonic origin of hematopoietic stem cells, established the dual origin of the endothelial network in the embryo and revealed the hematopoietic function of the allantois in birds, she has switched to mammals and contributed to demonstrating that the aorta and allantois/placenta are new sites of hematopoietic production in the mouse embryo. The manifold insights generated by the pivotal work of Françoise Dieterlen-Lièvre have created multiple paradigm shifts which continue to challenge the field of developmental hematopoiesis.

Allantois and placenta as developmental sources of hematopoietic stem cells.

While the aortic region, the para-aortic splanchnopleura/aorta-gonads-mesonephros (P-Sp/AGM) is currently considered as the source of definitive hematopoietic stem cells during development, the mouse placenta has been found to generate large numbers of these cells and to remain functional in this respect for a longer period than the P-Sp/AGM. The fetal component, which derives from the fused allantois and chorion, is responsible for this activity. We and others have shown that the pre-fusion allantois (before the stage of 6 pairs of somites) is able to yield clonogenic progenitors, provided that it is pre-cultured in toto before it is dissociated into single cells and seeded in semi-solid medium. Thus placental hematopoiesis can be concluded to derive from intrinsic precursors. It is similar in this regard to the yolk sac which both produces hematopoietic progenitors and supports their multiplication and differentiation. Hematopoietic activity, detected by in vitro colony assays, has also been recently uncovered in the human placenta. According to the data available, this newly identified source probably provides a large number of HSC during development and must play a foremost role in founding the definitive hematopoietic system.

Decoding the hemogenic endothelium in mammals.

A collection of recent Nature papers examines the relationship between endothelial precursors and hematopoietic cells. Two of these studies (Eilken et al., 2009; Lancrin et al., 2009) use time-lapse imaging with live markers and genetic analysis of differentiating ESCs to reveal that even non-aortic-derived endothelial cells are hemogenic.

Emergence of haematopoietic stem cells during development.

Self-renewable haematopoietic stem cells (HSCs) become segregated during development into a finite pool, from which they are mobilized upon physiological requirement. A central feature characterizing developmental haematopoiesis is that definitive organs become colonized by HSCs originating from a central source. The emission of HSCs occurs more or less continuously during a protracted period in parallel or successive sites. The most recently discovered of these sites is the placenta. The allantois, which is one of the components of the placenta, probed before it becomes vascularised, turns out to be a location where clonogenic precursors become committed. The placenta is thus a site of intrinsic haematopoiesis. Until this finding, the aorta and periaortic tissues were held to be the sites of definitive HSC commitment. The haematopoietic process in the aorta is prominent, particularly in avian embryos, and displays striking anatomical relationships between endothelial and haematopoietic cells. This made it possible to investigate the cytological and molecular relationship between the two types of cells. Somite exchanges between quail and chicken disclosed two distinct lineages, a dorsal one, purely endothelial, and a ventral one, hemangioblastic. The latter, also termed hemogenic endothelium, builds at first the whole inside lining of the aorta, and is then progressively replaced by cells of somitic origin, beginning with the aortic roof; it emits haematopoietic cells when located in the floor of the aorta and disappears. These events involve a changing molecular pattern, with expressions of transcription factor Runx1 and receptor VEGF-R2 as faithful markers of the lineage switch. Taking advantage of the stereotyped anatomical arrangement at the aortic level, which is favourable to dissect the mechanisms of HSC commitment, the analysis of developmental haematopoiesis should progress still further.

Hematopoietic potential of the pre-fusion allantois.

We previously showed that the fetal component of the placenta has a vigorous hematopoietic activity. Whether this organ is an environmental niche where hematopoietic stem cells (HSC) proliferate and become committed to various lineages, or whether it is also a site for HSC emergence, was left open. This issue can be addressed only if the components that will give rise to the placenta are tested prior to vascularization. The fetal part of the placenta forms through the fusion of the allantois and the chorionic plate around the stage of 7 somite pairs. The allantois, a mesodermal rudiment that provides fetal blood vessels to the placenta, was retrieved before fusion. We found in this rudiment expression of CD41, a known marker of early embryonic hematopoietic progenitors. c-Kit encoding a progenitor specific receptor was also expressed. Significantly, as early as the 1-2 somite stage, the allantois yielded erythroid, myeloid and multipotent clonogenic progenitors, when pre-cultured in toto prior to seeding in a semisolid medium. These results provide evidence that the allantois has hematopoietic potential per se. Whether this potential also involves the ability to produce HSC is still to be determined.

Are intra-aortic hemopoietic cells derived from endothelial cells during ontogeny?

The aorta is recognized as an intraembryonic site that produces adult-type hemopoietic stem cells. A corpus of data indicates that hemopoietic cells arranged as clusters attached to the aortic floor derive from an endothelial intermediate. This review reports on experimental approaches carried out in the avian embryo to establish the developmental history of the aortic endothelium and trace the origin of associated hemopoietic cells.

Placenta as a site for hematopoietic stem cell development.

The discovery of a major hematopoietic stem cell (HSC) pool in mid-gestation mouse placenta has defined the placenta as yet another important anatomical site that participates in HSC development. Placental HSC activity starts in parallel with the AGM region, before HSCs are found in circulation or have colonized the fetal liver. Moreover, placental hematopoietic activity culminates in a rapid expansion of the definitive HSC pool, which occurs during the time when the fetal liver HSC reservoir begins to grow. Furthermore, hematopoietic cells in mid-gestation mouse placenta are not instructed for differentiation along the myeloerythroid lineage, as in the fetal liver. These findings suggest that the placenta provides a supportive niche where the definitive hematopoietic stem cell pool can be temporarily established during development. Future studies are needed to characterize the developmental events that lead to the establishment of placental HSC pool, and to define the microenvironmental signals that support this process. Furthermore, if the stem cell-promoting properties of the placental niche can be harnessed in vitro to support HSC formation, maturation, and/or expansion in culture, these assets may greatly improve hematopoietic stem cell-based therapies in the future.

Commitment of hematopoietic stem cells in avian and mammalian embryos: an ongoing story.

During ontogeny, hematopoietic stem cells (HSC) become committed outside of hematopoietic organs. Once held to emerge from the yolk sac, they are currently thought to originate from the early aorta. However we now show that the allantois in the avian embryo and the placenta in the mouse embryo produce HSC in very large numbers. These sites thus appear to have a central role in the process of HSC emergence.

The placenta is a niche for hematopoietic stem cells.

The hematopoietic system develops during embryogenesis at temporally and anatomically restricted sites. The anatomical origin of definitive HSCs is not fully resolved, and little is known about how the different fetal hematopoietic microenvironments direct HSC development. Here, we show that the mouse placenta functions as a hematopoietic organ that harbors a large pool of pluripotent HSCs during midgestation. The onset of HSC activity in the placenta parallels that of the AGM (aorta-gonad-mesonephros) region starting at E10.5-E11.0. However, the placental HSC pool expands until E12.5-E13.5 and contains >15-fold more HSCs than the AGM. The expansion of the CD34(+)c-kit(+) HSC pool in the placenta occurs prior to and during the initial expansion of HSCs in the fetal liver. Importantly, the placental HSC pool is not explained by rare circulating HSCs, which appear later. These data support an important, but unappreciated, role for the placenta in establishing the mammalian definitive hematopoietic system.

In vivo methods to analyze cell origins, migrations, homing, and interactions in the blood, vascular, and immune systems of the avian and mammalian embryo.

In vivo experimental approaches that have been designed to study the ontogeny of the hematopoietic system in higher vertebrates are described in the present chapter. The avian embryo is directly available to manipulations in ovo during gastrulation and organogenesis. This permissiveness has led to the design of various approaches that provided crucial insights into the ontogeny of the hematopoietic system, particularly regarding traffic of progenitors between different compartments. In contrast, experimental manipulation of the developing mouse in utero is possible only during the second half of gestation, that is, the fetal period. This approach has been very useful in understanding how the immune system learns to distinguish self from nonself.

From the hemangioblast to self-tolerance: a series of innovations gained from studies on the avian embryo.

During the last decades of the 20th century, studies on the vertebrate hematopoietic and immune systems have largely been performed, on mammalian models. The mouse has been the preferred material for several cogent reasons: (i) numerous well defined genetic strains are available; (ii) this species has been and still is instrumental in the study of gene activity through transgenesis; and (iii) in vitro culture techniques and in vivo assays for blood cells together with a wide array of antibodies and nucleic acid probes have been developed to investigate the cellular interactions occurring during hematopoiesis and immune reactivity. However, important and fundamental notions have emerged from using another higher vertebrate model, the avian embryo. The distinction among small lymphocytes of two populations, the T and B lymphocytes, endowed with different roles in adaptive immunity and dependant on different environments for their specification, has relied on experiments carried out in birds. The avian model has been critical for the analysis of the origin and traffic of hematopoietic precursor cells. It allowed the demonstration that both hematopoietic and angioblastic lineages arise from a common precursor, a cell whose existence had been proposed but never undoubtedly proven, the hemangioblast. Finally a form of thymus-dependant 'dominant' tolerance was demonstrated on the basis of experiments in the avian embryo, which initiated a large current of studies on 'regulatory T-cells'. Work in this model during the last decades has relied strongly on the construction of chimeras between quail and chick embryos that allowed a refined analysis of cell behaviour during embryogenesis. The novel perception of developmental neuropoiesis and immunopoiesis that followed proved to be largely applicable to lower and higher vertebrates, notably mammals.

Mouse placenta is a major hematopoietic organ.

Placenta and yolk sac from 8- to 17-day-old (E8-E17) mouse embryos/fetuses were investigated for the presence of in vitro clonogenic progenitors. At E8-E9, the embryonic body from the umbilicus caudalwards was also analysed. Fetal liver was analysed beginning on E10. At E8, between five and nine somite pairs (sp), placenta, yolk sac and embryonic body yielded no progenitors. The first progenitors appeared at E8.5 at the stage of 15 sp in the yolk sac, 18 sp in the embryonic body, 20 sp in the placenta and only at E12 in the fetal liver (absent at E10, at E11 not determined). Progenitors with a high proliferation potential that could be replated for two months, as well as the whole range of myeloid progenitors, were found at all stages in all organs. However, the earliest of these progenitors (these yielding large, multilineage colonies) were 2-4 times more frequent in the placenta than in the yolk sac or fetal liver. In the fetal liver, late progenitors were more frequent and the cellularity increased steeply with developmental age. Thus, the fetal liver, which is a recognized site for amplification and commitment, has a very different hematopoietic developmental profile from placenta or yolk sac. Placentas were obtained from GFP transgenic embryos in which only the embryonic contribution expressed the transgene. 80% of the colonies derived from these placental cells were GFP+, and so originated from the fetal component of the placenta. These data point to the placenta as a major hematopoietic organ that is active during most of pregnancy.

Hemangioblasts and hemopoietic stem cells during ontogeny.

This review focuses on the emergence of hemopoietic stem cells (HSC) in the embryonic aorta, which was analysed in the avian model. Intraaortic clusters, a characteristic vertebrate anatomical feature, were shown to derive from the splanchnopleural (ventral) mesoderm, which has the potential to give rise to both angioblasts and hemopoietic cells. In contrast, the somitic mesoderm was shown to give rise to angioblasts only. The derivation of hemopoietic progenitors from endothelial cells in the floor of the aorta was followed by means of in vivo labelling experiments. Finally, the expression of gene-encoding transcription factors involved in the emergence of HSC was restricted to the floor of the aorta immediately prior to and during the appearance of intraaortic clusters.

[Exocrine enzymes in the pancreatic rudiments of the chick embryo developing in association with homologous or heterologous mesenchyme]. / Développement des enzymes exocrines dans les bourgeons pancréatiques chez l'embryon de poulet en présence de mésenchymes homologues et hétérologues.

The aim of the present study was to investigate the development of enzymes in the chick pancreas and to study the influence of a heterologous mesenchyme (lung mesenchyme) on the differentiation of these enzymes. 1. The normal development of chymotrypsin, carboxypeptidase A, amylase, and lipase in the chick pancreas has been studied. The proteolytic activities as well as the amylase become detectable between day 14 and day 17. The total and specific activities of these enzymes remain rather low during about a week, and they show a spectacular rise just before hatching. In contrast to the proteolytic and amylolytic enzymes, lipase could not be detected before day 20 and showed a rapid increase just before and during hatching. 2. These enzymes have also been studied in the spleen and in the lung. Both organs show low and constant activities of chymotrypsin, carboxypeptidase A, and amylase. They are devoid of lipase. The latter enzyme appears to be strictly specific for the pancreas. 3. The specific activities of the various enzymes were measured in the various lobes of the pancreas; however, no significant differences were found. 4. All enzymatic activities were found in the organs obtained from an association of pancreatic epithelium with pulmonary mesenchyme and which developed in coelomic grafts. These activities were as high as in control organs where the pancreatic epithelium was not dissociated from its own mesenchyme. In conclusion it was found that the determined pancreatic epithelium, if it is associated with a pulmonary mesenchyme at the early embryonic stage, gives rise to biochemically normal exocrine pancreatic tissue.