Definitive hematopoiesis initiates through a committed erythromyeloid progenitor in the zebrafish embryo

Shifting sites of blood cell production during development is common across widely divergent phyla. In zebrafish, like other vertebrates, hematopoietic development has been roughly divided into two waves, termed primitive and definitive. Primitive hematopoiesis is characterized by the generation of embryonic erythrocytes in the intermediate cell mass and a distinct population of macrophages that arises from cephalic mesoderm. Based on previous gene expression studies, definitive hematopoiesis has been suggested to begin with the generation of presumptive hematopoietic stem cells (HSCs) along the dorsal aorta that express c-myb and runx1. Here we show, using a combination of gene expression analyses, prospective isolation approaches, transplantation, and in vivo lineage-tracing experiments, that definitive hematopoiesis initiates through committed erythromyeloid progenitors (EMPs) in the posterior blood island (PBI) that arise independently of HSCs. EMPs isolated by coexpression of fluorescent transgenes driven by the lmo2 and gata1 promoters exhibit an immature, blastic morphology and express only erythroid and myeloid genes. Transplanted EMPs home to the PBI, show limited proliferative potential, and do not seed subsequent hematopoietic sites such as the thymus or pronephros. In vivo fate-mapping studies similarly demonstrate that EMPs possess only transient proliferative potential, with differentiated progeny remaining largely within caudal hematopoietic tissue. Additional fate mapping of mesodermal derivatives in mid-somitogenesis embryos suggests that EMPs are born directly in the PBI. These studies provide phenotypic and functional analyses of the first hematopoietic progenitors in the zebrafish embryo and demonstrate that definitive hematopoiesis proceeds through two distinct waves during embryonic development.     

A global role for zebrafish klf4 in embryonic erythropoiesis

There are two waves of erythropoiesis, known as primitive and definitive waves in mammals and lower vertebrates including zebrafish. The founding member of the Kruppel-like factor (KLF) family of CACCC-box binding proteins, EKLF/Klf1, is essential for definitive erythropoiesis in mammals but only plays a minor role in primitive erythropoiesis. Morpholino knockdown experiments have shown a role for zebrafish klf4 in primitive erythropoiesis and hatching gland formation. In order to generate a global understanding of how klf4 might influence gene expression and differentiation, we have performed expression profiling of klf4 morphants, and then performed validation of many putative target genes by qRT-PCR and whole mount in situ hybridization. We found a critical role for klf4 in embryonic globin, heme synthesis and hatching gland gene expression. In contrast, there was an increase in expression of definitive hematopoietic specific genes such as larval globin genes, runx1 and c-myb from 24 hpf, suggesting a selective role for klf4 in primitive rather than definitive erythropoiesis. In addition, we show klf4 preferentially binds CACCC box elements in the primitive zebrafish beta-like globin gene promoters. These results have global implications for primitive erythroid gene regulation by KLF-CACCC box interactions.     

Hemogenic endothelium: origins, regulation, and implications for vascular biology

The study of endothelial development has been intertwined with hematopoiesis since the early 20th century when a bi-potential cell (hemangioblast) was noted to produce both endothelial and hematopoietic cells. Since then, ideas regarding the nature of connection between the vascular and hematopoietic systems have ranged from a tenuous association to direct lineage origination. In this review, historical data that spans hematopoietic development is examined within the context of hemogenic endothelium. Hemogenic endothelium, a specialized endothelial population capable of hematopoiesis, is an emerging theory that has recently gained momentum. Evidence across species and decades are reviewed, as are the possible modulators of the phenomenon, which include pathways that specify definitive hematopoiesis (Runx1), arterial identity (Notch1), as well as physiological and developmental factors.     

Characterization of a weak allele of zebrafish cloche mutant

Hematopoiesis is a complicated and dynamic process about which the molecular mechanisms remain poorly understood. Danio rerio (zebrafish) is an excellent vertebrate system for studying hematopoiesis and developmental mechanisms. In the previous study, we isolated and identified a cloche(172) (clo(172)) mutant, a novel allele compared to the original cloche (clo) mutant, through using complementation test and initial mapping. Here, according to whole mount in-situ hybridization, we report that the endothelial cells in clo(172) mutant embryos, although initially developed, failed to form the functional vascular system eventually. In addition, further characterization indicates that the clo(172) mutant exhibited weaker defects instead of completely lost in primitive erythroid cells and definitive hematopoietic cells compared with the clo(s5) mutant. In contrast, primitive myeloid cells were totally lost in clo(172) mutant. Furthermore, these reappeared definitive myeloid cells were demonstrated to initiate from the remaining hematopoietic stem cells (HSCs) in clo(172) mutant, confirmed by the dramatic decrease of lyc in clo(172)runx1(w84x) double mutant. Collectively, the clo(172) mutant is a weak allele compared to the clo(s5) mutant, therefore providing a model for studying the early development of hematopoietic and vascular system, as well as an opportunity to further understand the function of the cloche gene.     

Induction of embryonic hematopoietic and endothelial stem/progenitor cells by hedgehog-mediated signals

Blood and vascular endothelial cells form in all vertebrates during gastrulation, a process in which the mesoderm of the embryo is induced and then patterned by molecules whose identity is still largely unknown. Blood islands' of primitive hematopoietic cell clusters surrounded by a layer of endothelial cells form in the yolk sac, external to the developing embryo proper. These lineages arise from a layer of extraembryonic mesoderm that is closely apposed with a layer of primitive (visceral) endoderm. Despite the identification of genes such as Flk1, SCL/tal-1, Cbfa2/Runx1/AML1 and CD34 that are expressed during the induction of primitive hematopoiesis and vasculogenesis, the early molecular and cellular events involved in these processes are not well understood. Recent work has demonstrated that extracellular signals secreted by visceral endoderm surrounding the embryo are essential for the initiation of these events. A member of the Hedgehog family of signaling molecules (Indian hedgehog) is produced by visceral endoderm, can induce formation of blood and endothelial cells in explant cultures and can reprogram prospective neurectoderm along hematopoietic and endothelial cell lineages. Hedgehog proteins also stimulate proliferation of definitive hematopoietic stem/progenitor cells. These findings may have important implications for regulating hematopoiesis and vascular development for therapeutic purposes in humans and for the development of new sources of stem cells for transplantation and gene therapy.     

The BMP ligand Gdf6 prevents differentiation of coronal suture mesenchyme in early cranial development

Growth Differentiation Factor-6 (Gdf6) is a member of the Bone Morphogenetic Protein (BMP) family of secreted signaling molecules. Previous studies have shown that Gdf6 plays a role in formation of a diverse subset of skeletal joints. In mice, loss of Gdf6 results in fusion of the coronal suture, the intramembranous joint that separates the frontal and parietal bones. Although the role of GDFs in the development of cartilaginous limb joints has been studied, limb joints are developmentally quite distinct from cranial sutures and how Gdf6 controls suture formation has remained unclear. In this study we show that coronal suture fusion in the Gdf6-/- mouse is due to accelerated differentiation of suture mesenchyme, prior to the onset of calvarial ossification. Gdf6 is expressed in the mouse frontal bone primordia from embryonic day (E) 10.5 through 12.5. In the Gdf6-/- embryo, the coronal suture fuses prematurely and concurrently with the initiation of osteogenesis in the cranial bones. Alkaline phosphatase (ALP) activity and Runx2 expression assays both showed that the suture width is reduced in Gdf6+/- embryos and is completely absent in Gdf6-/- embryos by E12.5. ALP activity is also increased in the suture mesenchyme of Gdf6+/- embryos compared to wild-type. This suggests Gdf6 delays differentiation of the mesenchyme occupying the suture, prior to the onset of ossification. Therefore, although BMPs are known to promote bone formation, Gdf6 plays an inhibitory role to prevent the osteogenic differentiation of the coronal suture mesenchyme.     

Rumba and Haus3 are essential factors for the maintenance of hematopoietic stem/progenitor cells during zebrafish hematopoiesis

The hallmark of vertebrate definitive hematopoiesis is the establishment of the hematopoietic stem/progenitor cell (HSPC) pool during embryogenesis. This process involves a defined ontogenic switching of HSPCs in successive hematopoietic compartments and is evolutionarily conserved from teleost fish to human. In zebrafish, HSPCs originate from the ventral wall of the dorsal aorta (VDA), from which they subsequently mobilize to an intermediate hematopoietic site known as the caudal hematopoietic tissue (CHT) and finally colonize the kidney for adult hematopoiesis. Despite substantial understanding of the ontogeny of HSPCs, the molecular basis governing migration, colonization and maintenance of HSPCs remains to be explored fully. Here, we report the isolation and characterization of two zebrafish mutants, rumba(hkz1) and samba(hkz2), that are defective in generating definitive hematopoiesis. We find that HSPC initiation in the VDA and subsequent homing to the CHT are not affected in these two mutants. However, the further development of HSPCs in the CHT is compromised in both mutants. Positional cloning reveals that Rumba is a novel nuclear C2H2 zinc-finger factor with unknown function and samba encodes an evolutionarily conserved protein that is homologous to human augmin complex subunit 3 (HAUS3). Furthermore, we show that these two factors independently regulate cell cycle progression of HSPCs and are cell autonomously required for HPSC development in the CHT. Our study identifies Rumba and Haus3 as two essential regulators of HSPC maintenance during zebrafish fetal hematopoiesis.