Abstract

The first definitive blood cells during embryogenesis are derived from endothelial cells in a highly conserved process known as endothelial-to-haematopoietic transition (EHT). This conversion involves activation of a haematopoietic transcriptional programme in a subset of endothelial cells in the major vasculature of the embryo, followed by major morphological changes that result in transitioning cells rounding up, breaking the tight junctions to neighbouring endothelial cells and adopting a haematopoietic fate. The whole process is co-ordinated by a complex interplay of key transcription factors and signalling pathways, with additional input from surrounding tissues. Diverse model systems, including mouse, chick and zebrafish embryos as well as differentiation of pluripotent cells in vitro, have contributed to the elucidation of the details of the EHT, which was greatly accelerated in recent years by sophisticated live imaging techniques and advances in transcriptional profiling, such as single-cell RNA-Seq. A detailed knowledge of these developmental events is required in order to be able to apply it to the generation of haematopoietic stem cells from pluripotent stem cells in vitro — an achievement which is of obvious clinical importance. The aim of this review is to summarise the latest findings and describe how these may have contributed towards achieving this goal.

Introduction

In vertebrate animals, a close relationship exists between haematopoietic cells and endothelial cells. Haematopoietic cells travel through endothelium-lined blood vessels; they migrate in and out of the blood stream via docking to and transversing the endothelial layer; and haematopoietic stem and progenitor cells (HSPCs) are regulated and maintained by a vascular niche in the bone marrow during adult life and in the foetal liver during foetal life [13]. However, the relationship could not be any more intimate than during the generation of the first definitive HSPCs since these are directly produced from endothelial precursors in a process termed endothelial-to-haematopoietic transition (EHT). This cell fate transition involves switching on a haematopoietic transcriptional programme in select endothelial cells, which then become haemogenic, followed by considerable morphological changes that lead to the breaking of tight junctions with neighbouring endothelial cells and rounding up, to then being released into the blood stream (Figure 1).

Stepwise formation of HSPCs from endothelial cells.

Figure 1.
Stepwise formation of HSPCs from endothelial cells.

Depicted are the morphological events thought to take place in the dorsal aorta of the mouse embryo during the formation of HSPCs. Up-regulation of a haematopoietic transcriptional programme in endothelial cells during the specification of HECs is marked by a light red colour. EHT is characterised by a shape change towards a rounder cell with further implementation of a haematopoietic programme and breaking of tight junctions with neighbouring endothelial cells. HECs giving rise to pro-HSCs and pre-HSCs is thought to result in the formation of intra-aortic clusters in which cells mature into haematopoietic stem and progenitor cells as they colonise the foetal liver.

Figure 1.
Stepwise formation of HSPCs from endothelial cells.

Depicted are the morphological events thought to take place in the dorsal aorta of the mouse embryo during the formation of HSPCs. Up-regulation of a haematopoietic transcriptional programme in endothelial cells during the specification of HECs is marked by a light red colour. EHT is characterised by a shape change towards a rounder cell with further implementation of a haematopoietic programme and breaking of tight junctions with neighbouring endothelial cells. HECs giving rise to pro-HSCs and pre-HSCs is thought to result in the formation of intra-aortic clusters in which cells mature into haematopoietic stem and progenitor cells as they colonise the foetal liver.

The ontogeny of the haematopoietic system is characterised by multiple waves, multiple origins and shifting locations. The murine yolk sac gives rise to an initial wave of primitive erythroid cells, starting from embryonic day (E) 7.5, inside structures known as yolk sac blood islands, which consist of the first population of endothelial cells surrounding the newly formed primitive blood cells. The simultaneous appearance of blood and endothelial cells from a seemingly homogeneous cluster of mesodermal precursors initially led to the theory that primitive haematopoiesis did not originate from haemogenic endothelial cells (HECs), but from a developmentally earlier precursor termed haemangioblast, which displayed clonal bipotentiality for blood and endothelium and which had been detected in vitro [4] and in vivo [5]. More recent studies, however, have questioned the existence of such a bipotential precursor and have instead suggested that primitive blood may also derive from HECs located in the yolk sac [6,7].

In contrast with primitive haematopoiesis, definitive blood emerges in the context of a fully formed and functional vascular system. This again initiates in the yolk sac with the appearance of erythroid-myeloid progenitors (EMPs) by E8.5 [8] and immune-restricted progenitors [9]: two progenitor types that seem to complement each other with their differentiation potential. Adult-type haematopoietic stem cells (HSCs) that can fully repopulate the haematopoietic system of irradiated adult recipients upon direct transfer are not detected until E10.5. They first emerge in the dorsal aorta within the aorta-gonads-mesonephros (AGM) region and its associated major vessels, the umbilical and vitelline arteries [1012]. Subsequently, they are also detected in the yolk sac, the placenta and the embryonic head, before they go on to colonise the foetal liver [1315]. In each case, HSCs are found in the major vasculature within these tissues; however, it is not clear whether they are formed de novo in all of these tissues and whether a similar process of EHT is taking place. For example, even though embryonic head endothelial cells were shown to give rise to blood cells that contribute to the adult haematopoietic system [14], endothelium-associated haematopoietic cell clusters, which are signs of active EHT, have never been observed in the head [16].

Because of its importance as the first site of HSC generation, the process of EHT and its underlying molecular mechanisms have been almost exclusively studied in the dorsal aorta, which is what this review will therefore also concentrate on. Many excellent reviews on the haemogenic endothelium have been published in recent years [1719], which this review will aim to provide an update on.

First experimental evidence

The concept of an endothelial origin for definitive blood is not new. Histological observations made already 100 years ago proposed the existence of haemogenic endothelium [2022]; however, experimental support came much later, initially in birds, but then also in mice. Using the avian system, which allows ex vivo manipulations, in this case, dye-labelling of endothelial cells prior to haematopoietic development, Jaffredo, Dieterlen-Lièvre and colleagues were able to demonstrate that the label was subsequently passed on to emerging blood cells [23]. Ten years later, temporally restricted genetic labelling allowed similar fate mapping in mice, which confirmed that endothelial cells gave rise to emerging blood cells in the dorsal aorta which then went on to colonise the foetal liver and the adult bone marrow, where they contributed to haematopoiesis long-term [24].

Advances in live imaging have allowed real-time observation of EHT [2528]. Zebrafish embryos, which are transparent and develop externally, lend themselves particularly well to such observational studies. Stunning movies from the Herbomel lab of live zebrafish embryos revealed how endothelial cells with a flat morphology initiate EHT by rounding up, pulling neighbouring endothelial cells towards each other and eventually pinching off as round haematopoietic cells that are released into the circulation [28].

How these images can be reconciled with intra-aortic cluster formation is currently unknown. Marker expression analysis in mouse embryos suggests that emerging blood cells undergo further maturation within these clusters as they lose endothelial properties and up-regulate a haematopoietic programme. Indeed, an elegant explant culture system developed in the Medvinsky lab was able to dissect EHT into intermediate steps [2931]. HECs were shown to mature into fully functional HSCs via a pro-HSC, pre-HSC I and pre-HSC II stage, which was accompanied by sequential up-regulation of the cell surface markers CD41, CD43 and CD45.

Recreating EHT in vitro

A long-term goal, which could be considered the Holy Grail of the field due to its obvious clinical significance, is the generation of fully functional HSCs from pluripotent stem cells (PSCs) in vitro. While robust protocols exist for the differentiation of mouse and human PSCs [embryonic stem cells/induced PSCs (iPSCs)] into definitive haematopoietic cells, it is thought that these recapitulate the processes taking place in the early yolk sac and do not lead to the generation of HSCs. Indeed, differentiation of PSCs into considerable numbers of HSCs that are capable of long-term, high-level, multi-lineage haematopoietic reconstitution upon direct transplantation has proven to be a major challenge, although significant advances have been made recently [32,33].

What has become clear from the many attempts is the need for a thorough understanding of how HSCs are generated in vivo during embryogenesis, which has sparked renewed interest into the properties of HECs and the inner workings of the EHT (recently reviewed in refs. [3436]). In fact, the PSC differentiation system with its accessibility and ease for manipulation has greatly contributed to our increased knowledge of the molecular details of the EHT.

Another approach for generating HSCs via recreating EHT in vitro that has shown great promise is the induction of a haemogenic programme in adult endothelial cells cultured within a vascular niche, which resulted in the production of HSCs with high-level, multi-lineage engraftment potential [37]. This study highlighted the fact that endothelial cells not only act as direct precursors of HSCs but are also components of the niche that supports the generation and maintenance of HSCs.

Transcription factors—complexes, levels and dynamics

Conversion of PSCs or adult ECs into HSCs has relied on the ectopic expression of key transcription factors that drive the endothelial-to-haematopoietic conversion by suppressing the endothelial transcriptional programme and instead switching on haematopoietic genes (Figure 2). Two key transcription factors that have been shown to be absolutely required for the generation of HSCs from HECs in vivo are Runx1 [28,3840] and Gata2 [4144]. Due to the technical difficulties associated with low cell numbers in developing embryos, little is currently known about the complexes and networks that these two transcription factors function in, although it was previously reported that two downstream targets of Runx1, the transcriptional repressors Gfi1 and Gfi1b, are also essential for successful completion of the EHT through repression of the endothelial programme via the recruitment of Kdm1a [45]. It is also known that the activity of Runx1 is absolutely dependent on the presence of the cofactor Cbfβ, with the knockout of Cbfb displaying an almost identical phenotype to Runx1 knockout embryos [46,47].

The regulation of EHT.

Figure 2.
The regulation of EHT.

Key transcription factors involved in the EHT are shown in red and key signalling pathways in green. Cell surface markers characteristic for each stage are shown in purple. Whether a gene is up-regulated or down-regulated during the process is highlighted by its position above or below the schematic, respectively.

Figure 2.
The regulation of EHT.

Key transcription factors involved in the EHT are shown in red and key signalling pathways in green. Cell surface markers characteristic for each stage are shown in purple. Whether a gene is up-regulated or down-regulated during the process is highlighted by its position above or below the schematic, respectively.

Considering the importance of Runx1 and Gata2 in the EHT, it may not come as a surprise that the timing and the levels of their expression need to be tightly controlled. Early indications came from studies showing that haploinsufficiency of Runx1 led to a temporal and spatial shift in HSC appearance [48], while half a dose of Gata2 resulted in reduced production of HSCs specifically in the AGM region [43]. Furthermore, it has been reported that HECs express lower levels of Runx1 than are expressed by haematopoietic progenitors [49].

Two Runx1 isoforms have been detected, with the Runx1b isoform being controlled by the proximal promoter P2, while the expression of the Runx1c isoform is under the control of the distal promoter P1. P2, and thus Runx1b, activity initiates earlier, during the initial stages of the EHT, while P1/Runx1c are activated later during the EHT and in already committed haematopoietic progenitors and remain more prominent in the adult haematopoietic system [50,51]. In addition, it was recently demonstrated, using the PSC system, that Runx1b levels need to be kept low for initiation and successful completion of the EHT, since increased levels of Runx1b accelerate the onset of EHT, but are unable to drive haematopoietic maturation [52].

Increased expression of Runx1 was also detected in embryos deficient for the Notch1 target genes Hes1 and Hes5 [53]. Despite these embryos containing larger intra-aortic clusters and increased numbers of phenotypic haematopoietic progenitors in the AGM, these were non-functional, and haematopoietic development was severely impaired. In this case, however, the authors linked the haematopoietic defect to a failure to down-regulate Gata2, which is normally repressed by Hes1. This study demonstrated that Gata2 levels also need to be tightly and temporally regulated, as a Notch1-induced Gata2 expression spike is essential for haematopoietic development, but sustained Gata2 expression renders haematopoietic progenitors non-functional.

Even though Notch1 signalling is required to initiate expression of important haematopoietic regulators such as Gata2, its activity also needs to be restricted to allow for the completion of the EHT, as high Notch activity promotes an arterial programme in endothelial cells, while suppressing the haematopoietic programme [54]. In fact, a reporter system that differentially labels cells with high or low Notch activity has identified two populations in the dorsal aorta, an arterial-restricted population with high Notch activity and a haematopoietic-competent population with low Notch activity [55]. That same study further revealed the involvement of two different Notch ligands, with Dll4 inducing high Notch activity and thus implementing the arterial programme, while Jag1 facilitated the haematopoietic programme by restricting Notch activity. A down-regulation of Notch signalling was also observed during the maturation of pre-HSCs into fully competent HSCs [56].

Another transcriptional regulator with a function analogous to Notch is Sox17. It is known to determine arterial fate [57], but also to specify HECs [58]. Like Notch, it promotes the endothelial programme over the haematopoietic programme and needs to be down-regulated for the EHT to proceed [54]. It was found to be co-expressed with Runx1 in HECs and has been suggested to repress Runx1 function [54,58,59]. Indeed, relative levels of Runx1 and Sox17 as determined by measuring the ratio of their mean fluorescence intensities by immunofluorescence microscopy and correlating this with morphological changes detected by scanning electron microscopy can identify cells at the different stages of EHT [59]. Flat and elongated endothelial cells that are firmly integrated into the endothelial layer had a very low Runx1:Sox17 ratio, whereas cells inside the intra-aortic clusters scored with a high Runx1:Sox17 ratio. Thus, the levels of Runx1 and Sox17 are negatively correlated, with an increasing ratio of Runx1:Sox17 associated with the adoption of a haematopoietic fate. Interestingly, this was also associated with the appearance of membrane protrusions, the significance of which needs to be determined.

Very recently, Gata2 levels during the EHT were reported to be highly dynamic as measured with the help of a short-lived fluorescent Venus reporter for Gata2 expression [60]. Not only were different levels of Gata2-Venus detected among endothelial cells and intra-aortic cluster cells, but the expression of the reporter in these cells demonstrated a highly dynamic and pulsatile nature that could be correlated with different stages along the EHT. Specifically, cells that were emerging from the endothelium to form intra-aortic clusters displayed a shorter periodicity of Gata2-Venus expression pulses than HECs, thus demonstrating a higher dynamic nature. In addition, cells rounding up at the initiation of the EHT also demonstrated higher pulse amplitudes. The significance of these Gata2 expression pulses is currently unknown, although a functional role is suggested by an alteration in pulsatile behaviour in Gata2 heterozygous embryos that have reduced haematopoietic activity.

RNA modifications—another level of regulation

A new mode of transcriptional control that has received a lot of attention lately and also has an impact on haematopoietic cell generation in the AGM involves RNA modifications. The most common RNA modification in eukaryotes is the creation of N6-methyl-adenosine, which is laid down by RNA-specific methyltransferases, such as Mettl3 [61,62]. These modifications are recognised by specific reader proteins, which then target these RNA molecules for further processing, such as splicing, or degradation. Mettl3 was found to be expressed in endothelial cells, including HECs, and haematopoietic cells in both zebrafish and mouse embryos [63]. Furthermore, morpholino-mediated knock-down of mettl3 in zebrafish embryos resulted in decreased runx1 and cmyb expression and reduced numbers of HECs and HSPCs, which were further impaired in their haematopoietic differentiation. Interestingly, vascular development was normal, but endothelial cells maintained their endothelial identity and could not undergo the EHT. This was to a large part due to increased levels of Notch pathway components, whose mRNAs are normally targets of mettl3 and become N6-methylated at the time of EHT. Morpholino targeting of the N6-methyl-adenosine reader ythdf2, which targets methylated mRNAs for degradation, resulted in a similar phenotype and demonstrated that down-regulation of Notch activity for completion of EHT is, at least in part, achieved by modulation and degradation of notch1a mRNA via mettl3 and ythdf2. This pathway is conserved in mice [63,64].

Notch activity is also down-regulated by Gpr183 signalling-mediated ubiquitination and degradation of the Notch1 protein [65].

Molecular details of EHT

Despite having known about the existence of EHT in the generation of HSPCs for a considerable amount of time and having identified some of the key transcriptional regulators and signalling pathways, knowledge of the molecular details, especially the morphological changes and restructuring involved, is still very much limited. Many transcriptional profiling studies on non-HECs, HECs, cells undergoing EHT, intra-aortic cluster cells and HSPCs, some even at the single-cell level, have been performed in recent years with the aim of shedding more light on this process [49,6671] (Table 1).

Table 1
Phenotype of cell populations analysed by transcriptional profiling
Cell population Phenotype Stage Profiling Reference 
Non-HECs
HECs
HSPCs 
Runx1 + 23GFPCdh5+Ter119CD45CD41
Runx1 + 23GFP+Cdh5+Ter119CD45CD41
Runx1 + 23GFP+Cdh5+Ter119CD45CD41+ 
E8.5, E9.25, E10.5 Microarrays and scqRT-PCR [49
Non-HECs
HECs
HSCs
HPCs 
Ly6a-GFPCD31+ckit
Ly6a-GFP+CD31+ckit
Ly6a-GFP+CD31+ckit+
Ly6a-GFPCD31+ckit+ 
E10.5 Bulk RNA-Seq [70
GFP+ ECs
GFP ECs
GFP+ HCCs
GFP HCCs 
Ly6a-GFP+CD31+Cdh5+Esam+ckit
Ly6a-GFPCD31+Cdh5+Esam+ckit
Ly6a-GFP+CD31+Cdh5+Esam+ckit+
Ly6a-GFPCD31+Cdh5+Esam+ckit+ 
E10.5 Bulk RNA-Seq [68
Trunk HECs
Trunk rest 
runx1-GFP+
runx1-GFP 
26–28 hpf Bulk RNA-Seq [67
ECs
Pre-HSC I
Pre-HSC II
HSC (FL)
HSC (FL) 
CD31+Cdh5+CD41CD43CD45Ter119
CD31+CD45CD41locKit+EPCRhi
CD31+CD45+CD41lo
LinSca1+Mac1loEPCR+
CD45+CD150+CD48EPCR+ 
E11.5
E11.5
E11.5
E12
E14 
10-cell and scRNA-Seq [71
HCCs
Non-HECs
HECs
EHT cells
Pre-HSC I
Pre-HSC II
HSPCs (YS) 
ckit+
Cdh5+Gfi1-tomatoGfi1b-GFPckit
Cdh5+Gfi1-tomato+Gfi1b-GFPckit
Cdh5+Gfi1-tomato+Gfi1b-GFPckit+
Cdh5+ckit+CD45
Cdh5+ckit+CD45+
ckit+ 
E11.5
E10/E11


E11

E10/E11 
scRNA-Seq [66
ECs-HSPCs
Non-HECs
HECs
Pre-HSPC I
Pre-HSPC II
Pro-HSCs
Pre-HSC I
Pre-HSC II
Non-HECs
HECs
Pre-HSPC I
Pre-HSPC II 
Cdh5+
Cdh5+CD44
Cdh5+CD44lockit
Cdh5+CD44lockit+
Cdh5+CD44hi
Cdh5+CD41+CD43CD45
Cdh5+CD41+CD43+CD45
Cdh5+CD41+CD43+CD45+
Cdh5+CD44
Cdh5+CD44lockit
Cdh5+CD44lockit+
Cdh5+CD44hi 
E10.5
E10



E10


E9.5/10/11

E9.5/10
E11 
scRNA-Seq
scqRT-PCR



scqRT-PCR


Bulk RNA-Seq 
[69
Cell population Phenotype Stage Profiling Reference 
Non-HECs
HECs
HSPCs 
Runx1 + 23GFPCdh5+Ter119CD45CD41
Runx1 + 23GFP+Cdh5+Ter119CD45CD41
Runx1 + 23GFP+Cdh5+Ter119CD45CD41+ 
E8.5, E9.25, E10.5 Microarrays and scqRT-PCR [49
Non-HECs
HECs
HSCs
HPCs 
Ly6a-GFPCD31+ckit
Ly6a-GFP+CD31+ckit
Ly6a-GFP+CD31+ckit+
Ly6a-GFPCD31+ckit+ 
E10.5 Bulk RNA-Seq [70
GFP+ ECs
GFP ECs
GFP+ HCCs
GFP HCCs 
Ly6a-GFP+CD31+Cdh5+Esam+ckit
Ly6a-GFPCD31+Cdh5+Esam+ckit
Ly6a-GFP+CD31+Cdh5+Esam+ckit+
Ly6a-GFPCD31+Cdh5+Esam+ckit+ 
E10.5 Bulk RNA-Seq [68
Trunk HECs
Trunk rest 
runx1-GFP+
runx1-GFP 
26–28 hpf Bulk RNA-Seq [67
ECs
Pre-HSC I
Pre-HSC II
HSC (FL)
HSC (FL) 
CD31+Cdh5+CD41CD43CD45Ter119
CD31+CD45CD41locKit+EPCRhi
CD31+CD45+CD41lo
LinSca1+Mac1loEPCR+
CD45+CD150+CD48EPCR+ 
E11.5
E11.5
E11.5
E12
E14 
10-cell and scRNA-Seq [71
HCCs
Non-HECs
HECs
EHT cells
Pre-HSC I
Pre-HSC II
HSPCs (YS) 
ckit+
Cdh5+Gfi1-tomatoGfi1b-GFPckit
Cdh5+Gfi1-tomato+Gfi1b-GFPckit
Cdh5+Gfi1-tomato+Gfi1b-GFPckit+
Cdh5+ckit+CD45
Cdh5+ckit+CD45+
ckit+ 
E11.5
E10/E11


E11

E10/E11 
scRNA-Seq [66
ECs-HSPCs
Non-HECs
HECs
Pre-HSPC I
Pre-HSPC II
Pro-HSCs
Pre-HSC I
Pre-HSC II
Non-HECs
HECs
Pre-HSPC I
Pre-HSPC II 
Cdh5+
Cdh5+CD44
Cdh5+CD44lockit
Cdh5+CD44lockit+
Cdh5+CD44hi
Cdh5+CD41+CD43CD45
Cdh5+CD41+CD43+CD45
Cdh5+CD41+CD43+CD45+
Cdh5+CD44
Cdh5+CD44lockit
Cdh5+CD44lockit+
Cdh5+CD44hi 
E10.5
E10



E10


E9.5/10/11

E9.5/10
E11 
scRNA-Seq
scqRT-PCR



scqRT-PCR


Bulk RNA-Seq 
[69

Abbreviations: EC, endothelial cell; EHT, endothelial-to-haematopoietic transition; FL, foetal liver; HEC, haemogenic endothelial cell; HCC, haematopoietic cluster cell; HPC, haematopoietic progenitor cell; hpf, hours post fertilisation; HSC, haematopoietic stem cell; HSPC, haematopoietic stem/progenitor cell; sc, single cell; YS, yolk sac.

One of the first profiling studies of the EHT took advantage of the previously identified Runx1 + 23 enhancer, which is active in definitive HSPCs and HECs and, when linked to a GFP reporter, serves as an excellent tool for the isolation of these populations [49,50,72]. Through a combination of expression analyses and functional assays, it was established that HECs gradually lose their endothelial identity between E8.5 and E10.5 as their haematopoietic potential increases, with clonogenic assays indicating that endothelial and haematopoietic capacities are mutually exclusive [49]. The initiation of the haematopoietic programme in endothelial cells destined to become HECs is co-ordinated by Runx1 and the rapid decline of Runx1 expression in endothelial cells after E10.5 points to a very transient nature of HECs.

Other studies have focused on one or two developmental time points and compared the different cell populations from non-HEC to HSPCs with the aim of identifying, and ordering along a differentiation trajectory, the intermediate cellular states with the accompanying transcriptional stages. The identification of additional cell surface markers [69,71] and the use of further fluorescent reporter mouse lines [66,68,70] have made these isolation strategies ever more sophisticated.

Using a combination of markers for endothelial cells (CD31), intra-aortic cluster cells (ckit) and HECs/HSPCs (Ly6a-GFP), non-HECs, HECs, HSCs and more mature haematopoietic cells were compared, which led to the identification of the G protein-coupled receptor Gpr56 as being up-regulated concomitant with the onset of the haematopoietic programme, with functional studies in zebrafish confirming a role for Gpr56 in HSC emergence [70]. This study also reported an up-regulation of genes in HECs associated with migration, Notch pathway activation and hypoxia. Indeed, the same group had previously described a role for the hypoxia-induced gene Hif1a in HSC development [73]. Interestingly, a cell cycle signature was not detected until the HSC state, suggesting that cell proliferation occurs after completion of haematopoietic differentiation [70].

Using the same Ly6a-GFP reporter line, but in combination with CD31, Cdh5, Esam and ckit, another group compared GFP+ and GFP− endothelial cells and intra-aortic cluster cells [68]. They described a striking up-regulation of inflammatory pathways in HECs that was not triggered by an infection. Specifically, they found that interferon signalling increased HSPC production. This novel role of sterile pro-inflammatory signals promoting HSPC emergence from HECs was also discovered by two other groups using the zebrafish system [67,74], who reported that TNFα produced by primitive neutrophils initiated an inflammatory response in HEC via Tnfr2 and the TLR4–MyD88–NFκB axis, with the Notch pathway being an important downstream target. The activation of this cascade was required for the specification of haematopoietic cells from HECs.

The advent of single-cell RNA-Seq has allowed a more detailed dissection of the differentiation pathway from endothelial cell to haematopoietic cell [66,69,71]. Additional intermediate cellular states have been uncovered that can be ordered along a differentiation trajectory using pseudotime algorithms, which is most prominently marked by a gradual down-regulation of endothelial-specific genes and an up-regulation of haematopoietic genes. Zhou et al. described EPCR as a cell surface marker up-regulated as endothelial cells differentiate into pre-HSCs [71], with another group describing a further enrichment for pre-HSCs through the combination of EPCR with Dll4 [75]. This differential expression of EPCR between endothelial cells and pre-HSCs allowed Zhou et al. to isolate and sequence single endothelial cells, type I and II pre-HSCs and foetal liver HSCs [71]. They noticed the biggest transcriptional change taking place during the transition from endothelial cell to pre-HSC, with a general increase in transcriptional activity in pre-HSCs, which included genes involved in mitochondrial aerobic respiration, in the mTOR pathway and also inflammation-related genes. In addition, a high degree of heterogeneity in cell cycle status in the pre-HSC population was observed. Indeed, it was recently described that intra-aortic cluster cells at an earlier maturation stage, located at the base of the cluster, are more quiescent than more mature cells found at the tip of the cluster which are more actively cycling [76]. This suggests an expansion of pre-HSCs prior to their colonisation of the foetal liver and maturation into HSCs. The same authors have subsequently mined their data sets specifically for long non-coding RNAs and detected an up-regulation of H19 at the pre-HSC stage. An endothelial-specific knockout of the differentially methylated region that controls H19 expression led to severely impaired EHT and formation of HSPCs. Mechanistically, H19 was shown to regulate promoter methylation of key EHT transcription factors, such as Runx1, with increased methylation in its absence causing a down-regulation of Runx1 [77].

Single-cell RNA-Seq profiling by Baron et al. also highlighted that the greatest number of transcriptional changes takes place during the EHT, while there were only few genes differentially expressed during the transition from a non-HEC to an HEC [66]. The authors also carried out a careful analysis of the composition of the intra-aortic clusters and found these to contain type I and type II pre-HSCs as well as more committed HSPCs. This study has sequenced the most comprehensive repertoire of cell populations from endothelial cells to HSPCs to date. With gene sets specifying each step having been identified, this should serve as an important resource.

In the most recent single-cell RNA-Seq study, CD44 was identified as a marker that becomes detectable specifically in the transitioning population in which endothelial and haematopoietic programmes are co-expressed. Its levels further increase with the onset of ckit expression and mark cells that undergo morphological changes during the EHT. In fact, together with ckit and Cdh5, CD44 allowed an accurate isolation of cell stages from non-HEC to mature HSPCs. Furthermore, CD44 appears to have a functional role, as a disruption of the interaction with its ligand hyaluronan interfered with the EHT. It was further noted that compared with non-HEC, HECs were less proliferative and displayed a lower metabolic activity, but exhibited an increase in the expression of genes linked to lipid biosynthesis, Notch pathway activation and autophagy. An exit from the cell cycle may be necessary for a cell to undergo the EHT.

EHT vs. epithelial-to-mesenchymal transition

On the surface, EHT seems to share similarities with another major cell fate change — the epithelial-to-mesenchymal transition (EMT), and it was this obvious analogy that led to the adoption of the term EHT. Like EHT, EMT plays an essential role during development in tissue morphogenesis, but it is also a process involved in metastasis, when tumour cells gain the ability to leave their tissue-of-origin and seed other sites in the body. It is interesting in this context that it was recently reported that homophilic interactions of CD44 on breast cancer cells enhance the formation of clusters of these cells, which leads to increased metastatic potential [78]. It should be investigated whether CD44 may also be important for intra-aortic cluster formation during EHT and what the functional significance of this may be.

Another pathway that is shared by the EMT and EHT is Tgfb signalling [37,7983]. Activation of the Tgfb pathway was shown to promote the EMT, while its role in the EHT has been seemingly controversial. While some reports suggested that activation of Tgfb signalling promotes the EHT [81,82], others suggested that Tgfb activity needs to be down-regulated for the EHT to proceed [37,80,83]. While some of the controversies may be explained by the suggestion that the compound SB431542, commonly employed as an inhibitor of the Tgfb pathway [37,83], may have a more complex mode of action and can in fact increase phosphorylation and activation of Smad2/3 [82], which are downstream effectors of Tgfb signalling, other reports suggest that Tgfb signalling, like Notch activation, is essential for the specification of HECs, but may have to be down-regulated for the EHT to proceed [80]. In this context, the role of HDAC1 may also need further clarification. It is clear that it is required for HSPC specification from endothelial cells; however, in one study, it was reported to act by modulating the Tgfb pathway and Smad2/3 [82], while another group found it to activate Smad1/5, which led to ERK inhibition and the promotion of the haematopoietic programme [84]. A similar controversy surrounds the related Bmp signalling pathway, which may also have to be down-regulated during the final stages of HSPC formation [85].

Remaining questions

Our understanding of the details of the EHT has increased substantially in recent years (Figure 2), aided by the development of more sophisticated profiling and analysis tools. Despite this increase in knowledge, routine and efficient generation of HSCs from PSCs has still not been achieved although recent progress is promising. What are we still missing?

The more we learn about the process of HSPC specification, the more complex it turns out to be with many unanswered questions remaining. Haematopoietic progenitor generation from PSCs is much more successful, so how does it differ from producing HSCs? The embryo generates both stem and progenitor cells from HEC, but the former is a much rarer event, and we do not currently know how these two processes differ and why an HEC in one case may give rise to a stem cell, while in other cases, a more mature progenitor is produced. There is also evidence that there are different types of HECs [86], and that the process of the EHT may not be the same in every haematopoietic tissue [16]. Related to this is the question of how HECs are specified in the first place. While grafting experiments in avian embryos suggested a spatial segregation of haemogenic and non-haemogenic-fated endothelial precursors early on in development [87], it was more recently suggested that all nascent endothelial cells are competent to become HECs [49].

Even though there is very good evidence from endothelial fate mapping and endothelial-specific knockout studies (described above) that HSCs are generated from HECs, it could nevertheless be speculated that HSCs are in fact generated via a different mechanism—one that has so far not been recreated in the PSC system. Some of the markers that are employed for endothelial-specific targeting, such as Cdh5, continue to be expressed on HSCs, and temporal control of expression in vivo can be challenging [88]. Some staining results suggest that HSC precursors may also be located in the sub-aortic mesenchyme [30], and a knockout of Runx1 resulted in the accumulation of Runx1-LacZ+ cells in the sub-endothelial layer [89]. Whether these HSC precursors, upon maturation to HSCs, directly traverse the endothelial layer, temporarily integrate into the endothelial layer or adopt an endothelial fate before undergoing EHT, requires further investigation.

We also know relatively little about the surrounding supportive cells that make up the microenvironment and how they promote the EHT. It is clear that there is something special about the AGM microenvironment that allows HSC specification to occur in this part of the vasculature so efficiently. We are slowly starting to learn more about these AGM-derived supportive signals [90,91].

It is also clear that a simple activation of key transcription factors is not sufficient, but that the timing, co-expression, levels and dynamics need to be precisely controlled. These are all challenges that may have to be overcome before we can rely on a patient-specific supply of in vitro-generated HSCs for the clinic.

Perspectives

  • The EHT is the process through which haematopoietic stem and progenitor cells are first generated during development. A deciphering of the details would allow blood generation from pluripotent stem cells in vitro for cell replacement therapies.

  • Several of the key transcription factors and signalling pathways driving EHT have been identified. Intermediate cellular and transcriptional states are being elucidated through novel ex vivo culture systems and expression profiling.

  • More insight into the timing, relative levels and dynamics of key transcription factor expression as well as the contribution of the microenvironment need to be gained before robust generation of haematopoietic stem cells in vitro can be achieved.

Abbreviations

     
  • AGM

    aorta-gonads-mesonephros

  •  
  • EC

    endothelial cell

  •  
  • EHT

    endothelial-to-haematopoietic transition

  •  
  • EMPs

    erythroid-myeloid progenitors

  •  
  • EMT

    epithelial-to-mesenchymal transition

  •  
  • HCC

    haematopoietic cluster cell

  •  
  • HEC

    haemogenic endothelial cell

  •  
  • HPC

    haematopoietic progenitor cell

  •  
  • HSC

    haematopoietic stem cell

  •  
  • HSPC

    haematopoietic stem/progenitor cell

  •  
  • PSCs

    pluripotent stem cells

  •  
  • iPSCs

    induced PSCs

Acknowledgements

K.O. is currently a fellow of Cancer Research UK and has previous work on the development of the blood system funded by the Kay Kendall Leukaemia Fund and Bloodwise.

Competing Interests

The Author declares that there are no competing interests associated with this manuscript.

References

References
1
Khan
,
J.A.
,
Mendelson
,
A.
,
Kunisaki
,
Y.
,
Birbrair
,
A.
,
Kou
,
Y.
,
Arnal-Estape
,
A.
et al.  (
2016
)
Fetal liver hematopoietic stem cell niches associate with portal vessels
.
Science
351
,
176
180
2
Morrison
,
S.J.
and
Scadden
,
D.T.
(
2014
)
The bone marrow niche for haematopoietic stem cells
.
Nature
505
,
327
334
3
Wei
,
Q.
and
Frenette
,
P.S.
(
2018
)
Niches for hematopoietic stem cells and their progeny
.
Immunity
48
,
632
648
4
Kennedy
,
M.
,
Firpo
,
M.
,
Choi
,
K.
,
Wall
,
C.
,
Robertson
,
S.
,
Kabrun
,
N.
et al.  (
1997
)
A common precursor for primitive erythropoiesis and definitive haematopoiesis
.
Nature
386
,
488
493
5
Huber
,
T.L.
,
Kouskoff
,
V.
,
Fehling
,
H.J.
,
Palis
,
J.
and
Keller
,
G.
(
2004
)
Haemangioblast commitment is initiated in the primitive streak of the mouse embryo
.
Nature
432
,
625
630
6
Padron-Barthe
,
L.
,
Temino
,
S.
,
Villa del Campo
,
C.
,
Carramolino
,
L.
,
Isern
,
J.
and
Torres
,
M.
(
2014
)
Clonal analysis identifies hemogenic endothelium as the source of the blood-endothelial common lineage in the mouse embryo
.
Blood
124
,
2523
2532
7
Stefanska
,
M.
,
Batta
,
K.
,
Patel
,
R.
,
Florkowska
,
M.
,
Kouskoff
,
V.
and
Lacaud
,
G.
(
2017
)
Primitive erythrocytes are generated from hemogenic endothelial cells
.
Sci. Rep.
7
,
6401
8
McGrath
,
K.E.
,
Frame
,
J.M.
,
Fegan
,
K.H.
,
Bowen
,
J.R.
,
Conway
,
S.J.
,
Catherman
,
S.C.
et al.  (
2015
)
Distinct sources of hematopoietic progenitors emerge before HSCs and provide functional blood cells in the mammalian embryo
.
Cell Rep.
11
,
1892
1904
9
Boiers
,
C.
,
Carrelha
,
J.
,
Lutteropp
,
M.
,
Luc
,
S.
,
Green
,
J.C.
,
Azzoni
,
E.
et al.  (
2013
)
Lymphomyeloid contribution of an immune-restricted progenitor emerging prior to definitive hematopoietic stem cells
.
Cell Stem Cell.
13
,
535
548
10
de Bruijn
,
M.F.
,
Speck
,
N.A.
,
Peeters
,
M.C.
and
Dzierzak
,
E.
(
2000
)
Definitive hematopoietic stem cells first develop within the major arterial regions of the mouse embryo
.
EMBO J.
19
,
2465
2474
11
Medvinsky
,
A.
and
Dzierzak
,
E.
(
1996
)
Definitive hematopoiesis is autonomously initiated by the AGM region
.
Cell
86
,
897
906
12
Muller
,
A.M.
,
Medvinsky
,
A.
,
Strouboulis
,
J.
,
Grosveld
,
F.
and
Dzierzak
,
E.
(
1994
)
Development of hematopoietic stem cell activity in the mouse embryo
.
Immunity
1
,
291
301
13
Gekas
,
C.
,
Dieterlen-Lievre
,
F.
,
Orkin
,
S.H.
and
Mikkola
,
H.K.
(
2005
)
The placenta is a niche for hematopoietic stem cells
.
Dev. Cell.
8
,
365
375
14
Li
,
Z.
,
Lan
,
Y.
,
He
,
W.
,
Chen
,
D.
,
Wang
,
J.
,
Zhou
,
F.
et al.  (
2012
)
Mouse embryonic head as a site for hematopoietic stem cell development
.
Cell Stem Cell.
11
,
663
675
15
Ottersbach
,
K.
and
Dzierzak
,
E.
(
2005
)
The murine placenta contains hematopoietic stem cells within the vascular labyrinth region
.
Dev. Cell.
8
,
377
387
16
Li
,
Z.
,
Vink
,
C.S.
,
Mariani
,
S.A.
and
Dzierzak
,
E.
(
2016
)
Subregional localization and characterization of Ly6aGFP-expressing hematopoietic cells in the mouse embryonic head
.
Dev. Biol.
416
,
34
41
17
Gritz
,
E.
and
Hirschi
,
K.K.
(
2016
)
Specification and function of hemogenic endothelium during embryogenesis
.
Cell Mol. Life Sci.
73
,
1547
1567
18
Lacaud
,
G.
and
Kouskoff
,
V.
(
2017
)
Hemangioblast, hemogenic endothelium, and primitive versus definitive hematopoiesis
.
Exp. Hematol.
49
,
19
24
19
Swiers
,
G.
,
Rode
,
C.
,
Azzoni
,
E.
and
de Bruijn
,
M.F.
(
2013
)
A short history of hemogenic endothelium
.
Blood Cells Mol. Dis.
51
,
206
212
20
Emmel
,
V.E.
(
1916
)
The cell clusters in the dorsal aorta of mammalian embryos
.
Am. J. Anat.
19
,
401
421
21
Jordan
,
H.E.
(
1916
)
Evidence of hemogenic capacity of endothelium
.
Anat. Rec.
10
,
417
420
22
Sabin
,
F.R.
(
1920
)
Studies on the origin of blood vessels and of red blood corpuscles as seen in the living blastoderm of chicks during the second day of incubation
.
Contrib. Embryol.
9
,
213
262
23
Jaffredo
,
T.
,
Gautier
,
R.
,
Eichmann
,
A.
and
Dieterlen-Lievre
,
F.
(
1998
)
Intraaortic hemopoietic cells are derived from endothelial cells during ontogeny
.
Development
125
,
4575
4583
PMID:
[PubMed]
http://dev.biologists.org/content/125/22/4575.article-info
24
Zovein
,
A.C.
,
Hofmann
,
J.J.
,
Lynch
,
M.
,
French
,
W.J.
,
Turlo
,
K.A.
,
Yang
,
Y.
et al.  (
2008
)
Fate tracing reveals the endothelial origin of hematopoietic stem cells
.
Cell Stem Cell
3
,
625
636
25
Bertrand
,
J.Y.
,
Chi
,
N.C.
,
Santoso
,
B.
,
Teng
,
S.
,
Stainier
,
D.Y.
and
Traver
,
D.
(
2010
)
Haematopoietic stem cells derive directly from aortic endothelium during development
.
Nature
464
,
108
111
26
Boisset
,
J.C.
,
van Cappellen
,
W.
,
Andrieu-Soler
,
C.
,
Galjart
,
N.
,
Dzierzak
,
E.
and
Robin
,
C.
(
2010
)
In vivo imaging of haematopoietic cells emerging from the mouse aortic endothelium
.
Nature
464
,
116
120
27
Eilken
,
H.M.
,
Nishikawa
,
S.
and
Schroeder
,
T.
(
2009
)
Continuous single-cell imaging of blood generation from haemogenic endothelium
.
Nature
457
,
896
900
28
Kissa
,
K.
and
Herbomel
,
P.
(
2010
)
Blood stem cells emerge from aortic endothelium by a novel type of cell transition
.
Nature
464
,
112
115
29
Rybtsov
,
S.
,
Batsivari
,
A.
,
Bilotkach
,
K.
,
Paruzina
,
D.
,
Senserrich
,
J.
,
Nerushev
,
O.
et al.  (
2014
)
Tracing the origin of the HSC hierarchy reveals an SCF-dependent, IL-3-independent CD43(-) embryonic precursor
.
Stem Cell Rep.
3
,
489
501
30
Rybtsov
,
S.
,
Sobiesiak
,
M.
,
Taoudi
,
S.
,
Souilhol
,
C.
,
Senserrich
,
J.
,
Liakhovitskaia
,
A.
et al.  (
2011
)
Hierarchical organization and early hematopoietic specification of the developing HSC lineage in the AGM region
.
J. Exp. Med.
208
,
1305
1315
31
Taoudi
,
S.
,
Gonneau
,
C.
,
Moore
,
K.
,
Sheridan
,
J.M.
,
Blackburn
,
C.C.
,
Taylor
,
E.
et al.  (
2008
)
Extensive hematopoietic stem cell generation in the AGM region via maturation of VE-cadherin + CD45+ pre-definitive HSCs
.
Cell Stem Cell
3
,
99
108
32
Pearson
,
S.
,
Cuvertino
,
S.
,
Fleury
,
M.
,
Lacaud
,
G.
and
Kouskoff
,
V.
(
2015
)
In vivo repopulating activity emerges at the onset of hematopoietic specification during embryonic stem cell differentiation
.
Stem Cell Rep.
4
,
431
444
33
Sugimura
,
R.
,
Jha
,
D.K.
,
Han
,
A.
,
Soria-Valles
,
C.
,
da Rocha
,
E.L.
,
Lu
,
Y.F.
et al.  (
2017
)
Haematopoietic stem and progenitor cells from human pluripotent stem cells
.
Nature
545
,
432
438
34
Blaser
,
B.W.
and
Zon
,
L.I.
(
2018
)
Making HSCs in vitro: don't forget the hemogenic endothelium
.
Blood
132
,
1372
1378
35
Garcia-Alegria
,
E.
,
Menegatti
,
S.
,
Batta
,
K.
,
Cuvertino
,
S.
,
Florkowska
,
M.
and
Kouskoff
,
V.
(
2016
)
Emerging concepts for the in vitro derivation of murine haematopoietic stem and progenitor cells
.
FEBS Lett.
590
,
4116
4125
36
Slukvin
,
I.I.
(
2016
)
Generating human hematopoietic stem cells in vitro – exploring endothelial to hematopoietic transition as a portal for stemness acquisition
.
FEBS Lett.
590
,
4126
4143
37
Lis
,
R.
,
Karrasch
,
C.C.
,
Poulos
,
M.G.
,
Kunar
,
B.
,
Redmond
,
D.
,
Duran
,
J.G.B.
et al.  (
2017
)
Conversion of adult endothelium to immunocompetent haematopoietic stem cells
.
Nature
545
,
439
445
38
Chen
,
M.J.
,
Yokomizo
,
T.
,
Zeigler
,
B.M.
,
Dzierzak
,
E.
and
Speck
,
N.A.
(
2009
)
Runx1 is required for the endothelial to haematopoietic cell transition but not thereafter
.
Nature
457
,
887
891
39
Okuda
,
T.
,
van Deursen
,
J.
,
Hiebert
,
S.W.
,
Grosveld
,
G.
and
Downing
,
J.R.
(
1996
)
AML1, the target of multiple chromosomal translocations in human leukemia, is essential for normal fetal liver hematopoiesis
.
Cell
84
,
321
330
40
Wang
,
Q.
,
Stacy
,
T.
,
Binder
,
M.
,
Marin-Padilla
,
M.
,
Sharpe
,
A.H.
and
Speck
,
N.A.
(
1996
)
Disruption of the Cbfa2 gene causes necrosis and hemorrhaging in the central nervous system and blocks definitive hematopoiesis
.
Proc. Natl Acad. Sci. U.S.A.
93
,
3444
3449
41
de Pater
,
E.
,
Kaimakis
,
P.
,
Vink
,
C.S.
,
Yokomizo
,
T.
,
Yamada-Inagawa
,
T.
,
van der Linden
,
R.
et al.  (
2013
)
Gata2 is required for HSC generation and survival
.
J. Exp. Med.
210
,
2843
2850
42
Gao
,
X.
,
Johnson
,
K.D.
,
Chang
,
Y.I.
,
Boyer
,
M.E.
,
Dewey
,
C.N.
,
Zhang
,
J.
et al.  (
2013
)
Gata2 cis-element is required for hematopoietic stem cell generation in the mammalian embryo
.
J. Exp. Med.
210
,
2833
2842
43
Ling
,
K.W.
,
Ottersbach
,
K.
,
van Hamburg
,
J.P.
,
Oziemlak
,
A.
,
Tsai
,
F.Y.
,
Orkin
,
S.H.
et al.  (
2004
)
GATA-2 plays two functionally distinct roles during the ontogeny of hematopoietic stem cells
.
J. Exp. Med.
200
,
871
882
44
Tsai
,
F.Y.
,
Keller
,
G.
,
Kuo
,
F.C.
,
Weiss
,
M.
,
Chen
,
J.
,
Rosenblatt
,
M.
et al.  (
1994
)
An early haematopoietic defect in mice lacking the transcription factor GATA-2
.
Nature
371
,
221
226
45
Thambyrajah
,
R.
,
Mazan
,
M.
,
Patel
,
R.
,
Moignard
,
V.
,
Stefanska
,
M.
,
Marinopoulou
,
E.
et al.  (
2016
)
GFI1 proteins orchestrate the emergence of haematopoietic stem cells through recruitment of LSD1
.
Nat. Cell Biol.
18
,
21
32
46
Niki
,
M.
,
Okada
,
H.
,
Takano
,
H.
,
Kuno
,
J.
,
Tani
,
K.
,
Hibino
,
H.
et al.  (
1997
)
Hematopoiesis in the fetal liver is impaired by targeted mutagenesis of a gene encoding a non-DNA binding subunit of the transcription factor, polyomavirus enhancer binding protein 2/core binding factor
.
Proc. Natl Acad. Sci. U.S.A.
94
,
5697
5702
47
Wang
,
Q.
,
Stacy
,
T.
,
Miller
,
J.D.
,
Lewis
,
A.F.
,
Gu
,
T.L.
,
Huang
,
X.
et al.  (
1996
)
The CBFbeta subunit is essential for CBFalpha2 (AML1) function in vivo
.
Cell
87
,
697
708
48
Cai
,
Z.
,
de Bruijn
,
M.
,
Ma
,
X.
,
Dortland
,
B.
,
Luteijn
,
T.
,
Downing
,
R.J.
et al.  (
2000
)
Haploinsufficiency of AML1 affects the temporal and spatial generation of hematopoietic stem cells in the mouse embryo
.
Immunity
13
,
423
431
49
Swiers
,
G.
,
Baumann
,
C.
,
O'Rourke
,
J.
,
Giannoulatou
,
E.
,
Taylor
,
S.
,
Joshi
,
A.
et al.  (
2013
)
Early dynamic fate changes in haemogenic endothelium characterized at the single-cell level
.
Nat. Commun.
4
,
2924
50
Bee
,
T.
,
Liddiard
,
K.
,
Swiers
,
G.
,
Bickley
,
S.R.
,
Vink
,
C.S.
,
Jarratt
,
A.
et al.  (
2009
)
Alternative Runx1 promoter usage in mouse developmental hematopoiesis
.
Blood Cells Mol. Dis.
43
,
35
42
51
Sroczynska
,
P.
,
Lancrin
,
C.
,
Kouskoff
,
V.
and
Lacaud
,
G.
(
2009
)
The differential activities of Runx1 promoters define milestones during embryonic hematopoiesis
.
Blood
114
,
5279
5289
52
Lie
,
A.L.M.
,
Marinopoulou
,
E.
,
Lilly
,
A.J.
,
Challinor
,
M.
,
Patel
,
R.
,
Lancrin
,
C.
et al.  (
2018
)
Regulation of RUNX1 dosage is crucial for efficient blood formation from hemogenic endothelium
.
Development
145
53
Guiu
,
J.
,
Shimizu
,
R.
,
D'Altri
,
T.
,
Fraser
,
S.T.
,
Hatakeyama
,
J.
,
Bresnick
,
E.H.
et al.  (
2013
)
Hes repressors are essential regulators of hematopoietic stem cell development downstream of notch signaling
.
J. Exp. Med.
210
,
71
84
54
Lizama
,
C.O.
,
Hawkins
,
J.S.
,
Schmitt
,
C.E.
,
Bos
,
F.L.
,
Zape
,
J.P.
,
Cautivo
,
K.M.
et al.  (
2015
)
Repression of arterial genes in hemogenic endothelium is sufficient for haematopoietic fate acquisition
.
Nat. Commun.
6
,
7739
55
Gama-Norton
,
L.
,
Ferrando
,
E.
,
Ruiz-Herguido
,
C.
,
Liu
,
Z.
,
Guiu
,
J.
,
Islam
,
A.B.
et al.  (
2015
)
Notch signal strength controls cell fate in the haemogenic endothelium
.
Nat. Commun.
6
,
8510
56
Souilhol
,
C.
,
Lendinez
,
J.G.
,
Rybtsov
,
S.
,
Murphy
,
F.
,
Wilson
,
H.
,
Hills
,
D.
et al.  (
2016
)
Developing HSCs become notch independent by the end of maturation in the AGM region
.
Blood
128
,
1567
1577
57
Corada
,
M.
,
Orsenigo
,
F.
,
Morini
,
M.F.
,
Pitulescu
,
M.E.
,
Bhat
,
G.
,
Nyqvist
,
D.
et al.  (
2013
)
Sox17 is indispensable for acquisition and maintenance of arterial identity
.
Nat. Commun.
4
,
2609
58
Clarke
,
R.L.
,
Yzaguirre
,
A.D.
,
Yashiro-Ohtani
,
Y.
,
Bondue
,
A.
,
Blanpain
,
C.
,
Pear
,
W.S.
et al.  (
2013
)
The expression of Sox17 identifies and regulates haemogenic endothelium
.
Nat. Cell Biol.
15
,
502
510
59
Bos
,
F.L.
,
Hawkins
,
J.S.
and
Zovein
,
A.C.
(
2015
)
Single-cell resolution of morphological changes in hemogenic endothelium
.
Development
142
,
2719
2724
60
Eich
,
C.
,
Arlt
,
J.
,
Vink
,
C.S.
,
Solaimani Kartalaei
,
P.
,
Kaimakis
,
P.
,
Mariani
,
S.A.
et al.  (
2018
)
In vivo single cell analysis reveals Gata2 dynamics in cells transitioning to hematopoietic fate
.
J. Exp. Med.
215
,
233
248
61
Jia
,
G.
,
Fu
,
Y.
and
He
,
C.
(
2013
)
Reversible RNA adenosine methylation in biological regulation
.
Trends Genet.
29
,
108
115
62
Liu
,
J.
,
Yue
,
Y.
,
Han
,
D.
,
Wang
,
X.
,
Fu
,
Y.
,
Zhang
,
L.
et al.  (
2014
)
A METTL3-METTL14 complex mediates mammalian nuclear RNA N6-adenosine methylation
.
Nat. Chem. Biol.
10
,
93
95
63
Zhang
,
C.
,
Chen
,
Y.
,
Sun
,
B.
,
Wang
,
L.
,
Yang
,
Y.
,
Ma
,
D.
et al.  (
2017
)
M(6)A modulates haematopoietic stem and progenitor cell specification
.
Nature
549
,
273
276
64
Lv
,
J.
,
Zhang
,
Y.
,
Gao
,
S.
,
Zhang
,
C.
,
Chen
,
Y.
,
Li
,
W.
et al.  (
2018
)
Endothelial-specific m(6)A modulates mouse hematopoietic stem and progenitor cell development via notch signaling
.
Cell Res.
28
,
249
252
65
Zhang
,
P.
,
He
,
Q.
,
Chen
,
D.
,
Liu
,
W.
,
Wang
,
L.
,
Zhang
,
C.
et al.  (
2015
)
G protein-coupled receptor 183 facilitates endothelial-to-hematopoietic transition via Notch1 inhibition
.
Cell Res.
25
,
1093
1107
66
Baron
,
C.S.
,
Kester
,
L.
,
Klaus
,
A.
,
Boisset
,
J.C.
,
Thambyrajah
,
R.
,
Yvernogeau
,
L.
et al.  (
2018
)
Single-cell transcriptomics reveal the dynamic of haematopoietic stem cell production in the aorta
.
Nat. Commun.
9
,
2517
67
He
,
Q.
,
Zhang
,
C.
,
Wang
,
L.
,
Zhang
,
P.
,
Ma
,
D.
,
Lv
,
J.
et al.  (
2015
)
Inflammatory signaling regulates hematopoietic stem and progenitor cell emergence in vertebrates
.
Blood
125
,
1098
1106
68
Li
,
Y.
,
Esain
,
V.
,
Teng
,
L.
,
Xu
,
J.
,
Kwan
,
W.
,
Frost
,
I.M.
et al.  (
2014
)
Inflammatory signaling regulates embryonic hematopoietic stem and progenitor cell production
.
Genes Dev.
28
,
2597
2612
69
Oatley
,
M.
,
Bölübasi
,
Ö.V
,
Svensson
,
V.
,
Shvartsman
,
M.
,
Ganter
,
K.
,
Zirngibl
,
K.
et al.  (
2018
)
Single-cell transcriptomics identifies CD44 as a new marker and regulator of haematopoietic stem cells development
.
bioRxiv
5
,
1
41
70
Solaimani Kartalaei
,
P.
,
Yamada-Inagawa
,
T.
,
Vink
,
C.S.
,
de Pater
,
E.
,
van der Linden
,
R.
,
Marks-Bluth
,
J.
et al.  (
2015
)
Whole-transcriptome analysis of endothelial to hematopoietic stem cell transition reveals a requirement for Gpr56 in HSC generation
.
J. Exp. Med.
212
,
93
106
71
Zhou
,
F.
,
Li
,
X.
,
Wang
,
W.
,
Zhu
,
P.
,
Zhou
,
J.
,
He
,
W.
et al.  (
2016
)
Tracing haematopoietic stem cell formation at single-cell resolution
.
Nature
533
,
487
492
72
Nottingham
,
W.T.
,
Jarratt
,
A.
,
Burgess
,
M.
,
Speck
,
C.L.
,
Cheng
,
J.F.
,
Prabhakar
,
S.
et al.  (
2007
)
Runx1-mediated hematopoietic stem-cell emergence is controlled by a gata/Ets/SCL-regulated enhancer
.
Blood
110
,
4188
4197
73
Imanirad
,
P.
,
Solaimani Kartalaei
,
P.
,
Crisan
,
M.
,
Vink
,
C.
,
Yamada-Inagawa
,
T.
,
de Pater
,
E.
et al.  (
2014
)
HIF1alpha is a regulator of hematopoietic progenitor and stem cell development in hypoxic sites of the mouse embryo
.
Stem Cell Res.
12
,
24
35
74
Espin-Palazon
,
R.
,
Stachura
,
D.L.
,
Campbell
,
C.A.
,
Garcia-Moreno
,
D.
,
Del Cid
,
N.
,
Kim
,
A.D.
et al.  (
2014
)
Proinflammatory signaling regulates hematopoietic stem cell emergence
.
Cell
159
,
1070
1085
75
Hadland
,
B.K.
,
Varnum-Finney
,
B.
,
Mandal
,
P.K.
,
Rossi
,
D.J.
,
Poulos
,
M.G.
,
Butler
,
J.M.
et al.  (
2017
)
A common origin for B-1a and B-2 lymphocytes in clonal pre-hematopoietic stem cells
.
Stem Cell Rep.
8
,
1563
1572
76
Batsivari
,
A.
,
Rybtsov
,
S.
,
Souilhol
,
C.
,
Binagui-Casas
,
A.
,
Hills
,
D.
,
Zhao
,
S.
et al.  (
2017
)
Understanding hematopoietic stem cell development through functional correlation of their proliferative status with the intra-aortic cluster architecture
.
Stem Cell Rep.
8
,
1549
1562
77
Zhou
,
J.
,
Xu
,
J.
,
Zhang
,
L.
,
Liu
,
S.
,
Ma
,
Y.
,
Wen
,
X.
et al.  (
2018
)
Combined single-cell profiling of lncRNAs and functional screening reveals that H19 Is pivotal for embryonic hematopoietic stem cell development
.
Cell Stem Cell
24
,
285
298.e5
78
Liu
,
X.
,
Taftaf
,
R.
,
Kawaguchi
,
M.
,
Chang
,
Y.F.
,
Chen
,
W.
,
Entenberg
,
D.
et al.  (
2019
)
Homophilic CD44 interactions mediate tumor cell aggregation and polyclonal metastasis in patient-derived breast cancer models
.
Cancer Discov.
9
,
96
113
79
Lamouille
,
S.
,
Xu
,
J.
and
Derynck
,
R.
(
2014
)
Molecular mechanisms of epithelial-mesenchymal transition
.
Nat. Rev. Mol. Cell Biol.
15
,
178
196
80
Lempereur
,
A.
,
Canto
,
P.Y.
,
Richard
,
C.
,
Martin
,
S.
,
Thalgott
,
J.
,
Raymond
,
K.
et al.  (
2018
)
The TGFbeta pathway is a key player for the endothelial-to-hematopoietic transition in the embryonic aorta
.
Dev. Biol.
434
,
292
303
81
Monteiro
,
R.
,
Pinheiro
,
P.
,
Joseph
,
N.
,
Peterkin
,
T.
,
Koth
,
J.
,
Repapi
,
E.
et al.  (
2016
)
Transforming growth factor beta drives hemogenic endothelium programming and the transition to hematopoietic stem cells
.
Dev. Cell.
38
,
358
370
82
Thambyrajah
,
R.
,
Fadlullah
,
M.Z.H.
,
Proffitt
,
M.
,
Patel
,
R.
,
Cowley
,
S.M.
,
Kouskoff
,
V.
et al.  (
2018
)
HDAC1 and HDAC2 modulate TGF-beta signaling during endothelial-to-hematopoietic transition
.
Stem Cell Rep.
10
,
1369
1383
83
Vargel
,
O.
,
Zhang
,
Y.
,
Kosim
,
K.
,
Ganter
,
K.
,
Foehr
,
S.
,
Mardenborough
,
Y.
et al.  (
2016
)
Activation of the TGFbeta pathway impairs endothelial to haematopoietic transition
.
Sci. Rep.
6
,
21518
84
Zhang
,
C.
,
Lv
,
J.
,
He
,
Q.
,
Wang
,
S.
,
Gao
,
Y.
,
Meng
,
A.
et al.  (
2014
)
Inhibition of endothelial ERK signalling by Smad1/5 is essential for haematopoietic stem cell emergence
.
Nat. Commun.
5
,
3431
85
Souilhol
,
C.
,
Gonneau
,
C.
,
Lendinez
,
J.G.
,
Batsivari
,
A.
,
Rybtsov
,
S.
,
Wilson
,
H.
et al.  (
2016
)
Inductive interactions mediated by interplay of asymmetric signalling underlie development of adult haematopoietic stem cells
.
Nat. Commun.
7
,
10784
86
Chen
,
M.J.
,
Li
,
Y.
,
De Obaldia
,
M.E.
,
Yang
,
Q.
,
Yzaguirre
,
A.D.
,
Yamada-Inagawa
,
T.
et al.  (
2011
)
Erythroid/myeloid progenitors and hematopoietic stem cells originate from distinct populations of endothelial cells
.
Cell Stem Cell.
9
,
541
552
87
Pardanaud
,
L.
,
Luton
,
D.
,
Prigent
,
M.
,
Bourcheix
,
L.M.
,
Catala
,
M.
and
Dieterlen-Lievre
,
F.
(
1996
)
Two distinct endothelial lineages in ontogeny, one of them related to hemopoiesis
.
Development
122
,
1363
1371
PMID:
[PubMed]
http://dev.biologists.org/content/122/5/1363.long
88
Senserrich
,
J.
,
Batsivari
,
A.
,
Rybtsov
,
S.
,
Gordon-Keylock
,
S.
,
Souilhol
,
C.
,
Buchholz
,
F.
et al.  (
2018
)
Analysis of Runx1 using induced gene ablation reveals Its essential role in Pre-liver HSC development and limitations of an In vivo approach
.
Stem Cell Rep.
11
,
784
794
89
North
,
T.
,
Gu
,
T.L.
,
Stacy
,
T.
,
Wang
,
Q.
,
Howard
,
L.
,
Binder
,
M.
et al.  (
1999
)
Cbfa2 is required for the formation of intra-aortic hematopoietic clusters
.
Development
126
,
2563
2575
PMID:
[PubMed]
http://dev.biologists.org/content/126/11/2563.long
90
McGarvey
,
A.C.
,
Rybtsov
,
S.
,
Souilhol
,
C.
,
Tamagno
,
S.
,
Rice
,
R.
,
Hills
,
D.
et al.  (
2017
)
A molecular roadmap of the AGM region reveals BMPER as a novel regulator of HSC maturation
.
J. Exp. Med.
214
,
3731
3751
91
Mirshekar-Syahkal
,
B.
,
Fitch
,
S.R.
and
Ottersbach
,
K.
(
2014
)
Concise review: From greenhouse to garden: the changing soil of the hematopoietic stem cell microenvironment during development
.
Stem Cells
32
,
1691
1700
This is an open access article published by Portland Press Limited on behalf of the Biochemical Society and distributed under the Creative Commons Attribution License 4.0 (CC BY).