CD32 captures committed haemogenic endothelial cells during human embryonic development

During embryonic development, blood cells emerge from specialized endothelial cells, named haemogenic endothelial cells (HECs). As HECs are rare and only transiently found in early developing embryos, it remains difficult to distinguish them from endothelial cells. Here we performed transcriptomic analysis of 28- to 32-day human embryos and observed that the expression of Fc receptor CD32 (FCGR2B) is highly enriched in the endothelial cell population that contains HECs. Functional analyses using human embryonic and human pluripotent stem cell-derived endothelial cells revealed that robust multilineage haematopoietic potential is harboured within CD32+ endothelial cells and showed that 90% of CD32+ endothelial cells are bona fide HECs. Remarkably, these analyses indicated that HECs progress through different states, culminating in FCGR2B expression, at which point cells are irreversibly committed to a haematopoietic fate. These findings provide a precise method for isolating HECs from human embryos and human pluripotent stem cell cultures, thus allowing the efficient generation of haematopoietic cells in vitro.

During embryonic development, blood cells emerge from specialized endothelial cells, named haemogenic endothelial cells (HECs).As HECs are rare and only transiently found in early developing embryos, it remains difficult to distinguish them from endothelial cells.Here we performed transcriptomic analysis of 28-to 32-day human embryos and observed that the expression of Fc receptor CD32 (FCGR2B) is highly enriched in the endothelial cell population that contains HECs.Functional analyses using human embryonic and human pluripotent stem cell-derived endothelial cells revealed that robust multilineage haematopoietic potential is harboured within CD32 + endothelial cells and showed that 90% of CD32 + endothelial cells are bona fide HECs.Remarkably, these analyses indicated that HECs progress through different states, culminating in FCGR2B expression, at which point cells are irreversibly committed to a haematopoietic fate.These findings provide a precise method for isolating HECs from human embryos and human pluripotent stem cell cultures, thus allowing the efficient generation of haematopoietic cells in vitro.
During embryonic development, cells continuously alter their characteristics to specify new fates and generate the entire spectrum of cell types and tissues.Sometimes cell differentiation involves abrupt lineage conversions as in the case of embryonic haematopoiesis.In fact, various experimental models have informed us that, during development, haematopoietic cells are produced from haemogenic endothelial cells (HECs), a specialized subpopulation of embryonic endothelium [1][2][3][4][5][6][7][8][9][10] .HECs are thought to generate blood cells via an endothelial-to-haematopoietic transition, which involves considerable transcriptional and morphological changes leading to the identity switch from endothelial cell to blood 1,2,[4][5][6][7]11 . HEC represent a central element of the distinct haematopoietic developmental programmes.In fact, HECs can be found in the different anatomical locations where haematopoiesis is observed, including the yolk sac (YS), Article https://doi.org/10.1038/s41556-024-01403-0 the DA (Fig. 1a).As such, ACE expression emerges as a potential human HEC marker in vivo, similarly to what is described for murine HECs 22 .
We then tested whether ACE can also distinguish subsets of endothelial cells in vitro and track haematopoietic potential in hPSC haematopoietic cultures.For this, we differentiated the human embryonic stem cell (hESC) line H1 using a method that specifies hPSCs into WNT-dependent (WNTd) NOTCH-dependent multipotent HOXA + HECs, indicative of intra-embryonic AGM-like haematopoiesis 14,[27][28][29] .In this setting, we observed that ACE is expressed by virtually all day 8 CD34 + cells, including 90 ± 1% of CD34 + CD43 neg CD73 neg CD184 neg cells that comprise HECs in day 8 WNTd hPSC haematopoietic cultures (Extended Data Fig. 1a,b) 14 .Since HECs represent only ~2% of WNTd CD34 + CD43 neg CD73 neg CD184 neg cells 14 , we concluded that ACE expression is unable to distinguish HECs from other vascular endothelial cells in these hPSC differentiating cultures.
Having established that ACE expression segregates with the haemogenic transcription factor RUNX1 in endothelial cells of the human embryonic DA, we used this surface marker to better characterize embryonic endothelial cells with haematopoietic potential and identify markers that track with HECs both in vivo and in vitro.For this, we isolated CD34 + CD45 neg cells on the basis of their ACE expression from the AGM region of four human embryos (E1, E2, E3 and E4) staged between 28 dpf and 32 dpf (CS12-CS13; Extended Data Fig. 1c) and performed whole-transcriptomic analysis using bulk RNA sequencing (RNA-seq).Principal component analysis (PCA) and unsupervised hierarchical clustering by k-means highlighted a clear segregation of CD34 + CD45 neg ACE + and CD34 + CD45 neg ACE neg (herein ACE + and ACE neg , respectively) transcriptomes (Fig. 1b,c and Extended Data Fig. 1d,e), with 785 differentially expressed genes (DEGs), of which 440 genes were upregulated and 345 were downregulated in ACE + (Fig. 1c and Supplementary Table 1).At the population level, both cell fractions expressed high levels of endothelial genes (CD34, CDH5, PECAM1 and TEK) while genes associated with venous fate (NR2F2 and FLRT2) had a lower expression in ACE + cells in comparison with ACE neg cells.In contrast, ACE + cells showed a higher expression of genes classically associated with arterial cells (BMX, GJA5, DLL4, CXCR4 and HEY2) and HECs (MYB, GFI1 and CD44; Fig. 1d).Overrepresentation analysis (ORA) and gene set enrichment analysis (GSEA) revealed that ACE + cells are positively associated with cell migration-and cell adhesion-related Gene Ontology (GO) terms, in line with the remodelling occurring during the emergence of blood cells from HECs 30 (Fig. 1e and Supplementary Tables 2-5).Conversely, ACE + cells are negatively associated with several cell cycle-related GO terms, suggesting that at a population level they are undergoing an active arterialization process that requires cell growth suppression 31 .These results indicated that, in the aortic endothelium of 5-week human embryos, ACE expression identifies cells showing an arterial gene signature as well as cells expressing genes pivotal for haematopoietic development.

CD32 is expressed in HECs in human embryos
To identify putative markers for the isolation of HECs, we focused our attention on DEGs coding for cell surface proteins showing higher expression in ACE + cells (Fig. 2a, Extended Data Fig. 2a and Supplementary Table 6).We found that FCGR2B, which encodes for an isoform of the Fc receptor CD32, ranks among the top ten cell-surface genes whose expression is enriched in ACE + cells.Given that CD32 is a marker of other specialized endothelial cells [32][33][34] and that Fc receptors are expressed together with endothelial markers in a subset of YS-derived haematopoietic progenitors 35 , we focused our attention on this gene.We analysed human embryo sections between 26 dpf and 30 dpf (CS12-CS13), which showed that CD32 is expressed together with CD34 in the aortic endothelial cells.CD32 + endothelial cells localized close to the bifurcation with the vitelline artery (VA) in the AGM region, a site known to contain a high frequency of haematopoietic clusters, and therefore potentially of HECs, at this stage (Fig. 2b) 25 .Remarkably, in the DA, CD32 where lineage-restricted haematopoietic progenitors are generated first, as well as the aorta-gonad-mesonephros (AGM), the site of the Notch-dependent haematopoietic stem cell (HSC) emergence [12][13][14] .As HECs represent a rare and transient population rapidly generating haematopoietic output, they have been difficult to characterize and therefore little is known about how the emergence of the haematopoietic lineage is orchestrated.
Traditionally, HECs have been isolated and characterized in animal models, using reporters under the control of the regulatory elements of the transcription factors that drive blood cell emergence, such as Runx1 and Gfi1 (refs.15-17), a strategy that cannot be used to study HECs in human embryos.Our previous data suggested angiotensin-converting enzyme (ACE, also known as CD143) as a potential marker of HSCs and their endothelial precursors in human embryo 18,19 .Recently, transcriptomic analyses have allowed the identification of putative HEC markers in both murine and human embryos, including ACE as well as CXCR4 and CD44 (refs.20-23), but these also enrich for arterial endothelial cells (AECs), anatomically associated with HECs 13,14,24 .This hinders the specific characterization of the unique endothelial population that generates blood cells and, consequently, the design of accurate protocols for the derivation of therapeutically relevant haematopoietic cells from human pluripotent stem cells (hPSCs).
In this Article, to overcome these limitations and identify broadly applicable cell-surface markers specific for human HECs both in vivo and in vitro, we performed transcriptomic analysis of endothelial populations displaying haematopoietic potential in the human embryo based on ACE expression.Here, we report that the Fc receptor CD32 is expressed on human embryonic endothelial cells with robust hemogenic potential.Likewise, CD32 allows for the identification of hPSC-derived HECs with higher specificity than other reported HEC markers.This provided an unprecedented opportunity to study how and when HECs initiate the haematopoietic programme.We show that HECs transit through different states and their haematopoietic commitment occurs before the expression of haematopoietic markers and cell cycle genes.In particular, CD32 marks a specific subpopulation of HECs that no longer require NOTCH activation and that are fully committed to haematopoiesis.The ability to capture this rare and rapid embryonic process allows the redefinition of how blood cells emerge in the embryo and will enable the efficient generation of haematopoietic cells in vitro for therapies.

ACE identifies distinct human embryonic endothelial subsets
We have previously shown that, within the AGM region of the human embryo, haematopoietic clusters first emerge in the dorsal aorta (DA) at 27 days post-fertilization (dpf; Carnegie Stage (CS) 12) 25 .Human haematopoietic clusters are located on the ventral side of the aortic endothelium and express CD34, a surface marker that they share with the surrounding endothelial cells 25,26 .Intra-aortic CD34 + haematopoietic clusters also express ACE, another marker that identifies cells with haematopoietic potential in the developing human embryo 18,19 .As ACE expression is also observed in endothelial cells within different haematopoietic sites during human development, including the DA 18,19 , we further analysed its expression pattern in the AGM region in parallel with the expression of the transcription factor RUNX1, which regulates the emergence of blood cells in the embryo 15,16 .Immunofluorescence analysis on human embryonic sections revealed that, at 23 dpf (CS10), ACE + cells can be found in the mesenchyme surrounding the DA, but not in the aortic wall (Fig. 1a).At this stage, RUNX1 expression cannot be observed in the human AGM region.However, at 27 dpf (CS12), when the first haematopoietic clusters emerge inside the DA, ACE expression is also detected on endothelial cells.This suggests that ACE + precursors might migrate from the subaortic mesenchyme to the DA, although the lineage relationship between ACE + mesenchymal cells and ACE + endothelial cells remains to be elucidated.Strikingly, ACE and RUNX1 expression co-localized in the endothelial cells lining the ventral site of Article https://doi.org/10.1038/s41556-024-01403-0marks both the intra-aortic haematopoietic clusters bordering on the ventral site and the underlying endothelial cells.Immunofluorescence analysis on consecutive sections showed that CD32 + endothelial cells in the DA co-express CD34, ACE and RUNX1 (Fig. 2c).In the CS12 human embryo, CD32 is also expressed together with CD34 in endothelial cells surrounding the haemogenic regions of the YS (Fig. 2d).This specific expression pattern at active haemogenic embryonic sites led us to functionally test whether CD32 identifies HECs.Thus, we isolated by fluorescence-activated cell sorting (FACS) the CD32 + and CD32 neg fractions of the CD34 + CD43 neg CD45 neg endothelial cell population containing HECs in the AGM and YS regions dissected from two CS13 embryos (E5 and E6) and assessed their potential to generate haematopoietic progenitors ex vivo (Fig. 2e and Extended Data Fig. 2b-d).The CD32 + endothelial fraction from both the AGM and YS regions of both

CD32 defines intra-embryonic hPSC-derived HECs
Based on our results obtained with the human embryos, we next investigated whether CD32 can be a reliable marker for isolating HECs from hPSC differentiating cultures.We monitored its expression in WNTd intra-embryonic-like HOXA + HECs (Extended Data Fig. 3a) 14,27,29,36 .Remarkably, CD32 expression identified only a small subset of the HEC-containing CD34 + CD43 neg CD73 neg CD184 neg DLL4 neg cell population, which was all ACE + (Fig. 3a,b and Extended Data Fig. 3b,c).In parallel, using the H9 hESC reporter line to evaluate the expression of RUNX1C, a RUNX1 isoform expressed on WNTd HECs as they begin to generate haematopoietic progenitors 14,27 , we observed that CD32 expression in   3c,d).To monitor whether this population can generate blood cells, we isolated CD32 + from CD34 + CD43 neg CD7 3 neg CD184 neg DLL4 neg day 8 WNTd cells and cultured them on Matrigel in the presence of haematopoietic cytokines known to promote and sustain haematopoietic differentiation (HEC culture), as we previously described 14 .Under these conditions, the cells formed an adhesive monolayer that generated haematopoietic progeny as demonstrated by the presence of round cells and of a population of RUNX1C-EGFP + and CD45 + cells 5 days later (Fig. 3e,f).As such, the CD32 + fraction displays the defining behaviour of bona fide HECs.
We therefore assessed the haematopoietic potential of CD34 + CD43 neg CD73 neg CD184 neg DLL4 neg CD32 +/neg (referred to as CD32 + and CD32 neg ) cell populations (Fig. 3b and Extended Data Fig. 3a,c).The CD32 + cells generated erythro-myeloid clonogenic progenitors with significantly higher frequency than CD32 neg cells in both H1 and H9 hESC lines (Fig. 3g), similar to what we showed in the human embryo (Fig. 2f).Given the residual haematopoietic potential observed in CD32 neg cells isolated from the AGM region and hPSC-derived haematopoietic cultures (Figs.2f and 3g), we isolated CD32 neg cells from day 8 WNTd haematopoietic cultures and tested whether they could generate CD32 + progeny harbouring haematopoietic potential.Indeed, the 2-day-long culture of CD32 neg cells gave rise to ~40% (39 ± 1%) of CD32 + cells (Extended Data Fig. 3d) that, when isolated, generated CD45 + haematopoietic cells in HEC cultures (Extended Data Fig. 3e).These data demonstrate that the CD32 neg cell population contains precursors of HECs that do not express CD32 yet.This suggests that the residual haematopoietic output observed in the CD32 neg fraction is due to a further maturation of CD32 neg into CD32 + HECs that will then generate haematopoietic cells.We then assessed lymphoid potential by analysing the defining lymphoid lineage for the WNTd haematopoietic programme, that is, T cells 36 .Upon 24 days of co-culture on OP9DLL4 stromal cells, only CD32 + cells could robustly generate CD4 + CD8 + T cells while the CD32 neg fraction did not display T cell potential under the same conditions (Fig. 3h,i).CD32 + -derived CD4 + CD8 + T cells could progress towards more mature stages and gave rise to CD3 + cells expressing TCRαβ or TCRγδ and the activation markers CD45RA, CD25 and CD27 (Extended Data Fig. 3f,g).Of note, in contrast to cord blood CD34 + blood progenitors able to generate only the definitive-restricted Vδ1 + γδ population, CD32 + HECs generated different subsets of γδ T cells, including the Vδ1 + population as well as the Vδ2 + subset that develops early in the human embryo (Extended Data Fig. 3h) 37 .In addition, CD32 + HECs displayed the potential to robustly generate CD45 + CD56 + natural killer (NK) cells (Extended Data Fig. 3i).Collectively, these results show that CD32 + cells represent a subpopulation endowed with robust multilineage intra-embryonic-like haematopoietic potential in hPSC cultures.

CD32 is a specific HEC marker
We next compared the specificity of CD32 as an hPSC-derived HEC marker against CD44, a surface marker often used to define HECs 20,21 .In day 8 WNTd CD34 + CD43 neg cells, CD44 is expressed by most DLL4 + cells (Extended Data Fig. 4a), in line with the reported CD44 expression in AECs 13,20 , while it distinguishes two subpopulations of CD34 + CD4 3 neg CD184 neg CD73 neg DLL4 neg cells (referred to as CD44 + and CD44 neg ) (Extended Data Fig. 4b).We next assessed the haemogenic potential of CD44 + and CD44 neg subpopulations and observed that the CD44 + fraction was enriched for HECs as it generates significantly more clonogenic progenitors than CD44 neg cells (Fig. 3j), in accordance with recent reports 20,21 .Given that both CD32 and CD44 expression enriches for cells with haemogenic potential, we analysed the relationship between these two markers in hPSC differentiating cultures.Since CD32 identifies a minor subpopulation of CD44 + cells (Extended Data Fig. 4c) in day 8 WNTd CD34 + CD43 neg CD184 neg CD73 neg DLL4 neg cells, we investigated if CD44 + HECs upregulate CD32 expression to give rise to haematopoietic cells.Kinetic analyses throughout HEC differentiation revealed that CD44 + CD32 neg isolated at day 8 of WNTd haematopoietic differentiation begin to express CD32 within 2 days of culture (Extended Data Fig. 4d).Given its expression dynamics, we hypothesized that CD32 might be a more specific marker for HECs than CD44.To compare the specificity of CD32 and CD44, we performed a single-cell HEC assay using CD32 + or CD44 + cells isolated at day 8 of WNTd haematopoietic cultures 14 .This clonal analysis revealed that the CD32 + subfraction was highly enriched for HECs, as 87.0 ± 4.0% of the cells that formed a clone (143/168) in the HEC assay generated exclusively CD45 + haematopoietic cells (Fig. 3k,l and Extended Data Fig. 4e,f).In stark contrast, the CD44 + fraction contained equal proportions of progenitors with either haematopoietic or non-haematopoietic potential (Fig. 3k,l).This single-cell analysis indicates that CD32 is a reliable marker for hPSC-derived WNTd HECs, as nearly all CD32 + cells harbour robust haematopoietic potential.In addition, the use of CD32 yields a significant improvement in enriching for HECs compared with CD44, a marker often used to identify human HECs.

CD32 identifies HECs in a specific NOTCH-independent state
To further characterize hPSC-derived CD32 + cells, we next asked if these cells have transcriptional similarity to HECs found in the developing human embryo.We analysed the transcriptomic profile of the CD32 + fraction sorted from day 8 WNTd haematopoietic cultures and, since HECs in the DA are found in close contact with non-haemogenic cells arterial cells 2 , we compared them with CD34 + CD43 neg CD184 + CD73 + DLL4 + cells (referred to as DLL4 + ) as control sample 14 .While DLL4 + cells displayed a significant enrichment for genes whose expression is associated with arterial fate in vivo (for example, CXCR4, DLL4, HEY1, HEY2, SOX17 and GJA5) 38 , CD32 + cells were enriched for the expression of genes associated with HECs and their haematopoietic progression in vivo, including RUNX1, GFI1, MYCN and RAB27B (Fig. 4a and Supplementary Table 7) 38 .
We next contrasted CD32 + and DLL4 + cells with human embryonic single-cell RNA sequencing (scRNA-seq) datasets publicly available 20,38 , which, integrated, comprise populations of AECs and HECs from AGM and YS regions at different developmental stages spanning from CS10 to CS16 (Fig. 4b).Since CD32 + cells express HOXA9 and HOXA10 similarly to AGM but in contrast to YS cells (Fig. 4c,d), we restricted our comparison to the datasets of CS10-CS16 AGM cells.While DLL4 + cells resemble more closely CS14-CS16 embryonic AECs, CD32 + cells exhibited strong similarity to HECs, regardless of the developmental stage (Fig. 4e), thus including those when HSC generation has been observed, that is, CS13-CS16.
We then investigated whether FCGR2B expression could further refine the identification of HECs using available human CS14-CS16 AGM scRNA-seq data 38 .In this dataset, while CDH5 + RUNX1 + PTPRC neg FCGR2B neg HECs display an enrichment for genes associated with arterial endothelium, the expression of genes characteristic of haematopoietic commitment segregates within CDH5 + RUNX1 + PTPRC neg FCGR2B + -expressing cells (Fig. 4f and Extended Data Fig. 5a).These data suggest that transcriptionally distinct states of HECs can be identified.
To assess HEC heterogeneity and interrogate whether FCGR2B expression could define a specific state of HECs, we performed scRNA-seq of day 8 WNTd CD34 + CD43 neg CD184 neg CD73 neg cells as they contain HECs as well as other endothelial progenitors and HEC precursors 14 .Unsupervised clustering revealed 22 transcriptionally distinct clusters which were mostly annotated to the major endothelial cell fates (Fig. 5a, Extended Data Fig. 5b and Supplementary Table 8).This scRNA-seq analysis confirmed that WNTd HECs showed transcriptional heterogeneity as a total of six clusters were enriched for cells differentially expressing RUNX1 (Fig. 5a, Extended Data Fig. 5b,c and Supplementary Table 8), including one with enriched expression of FCGR2B (cluster 11 in Fig. 5a, Extended Data Fig. 5d and Supplementary Table 8).Pseudotime analysis by Monocle3 revealed that FCGR2B + HECs represent an intermediate state for the progression of RUNX1 + KCNK17 + H19 + FCGR2B neg HECs 39 to RUNX1 + cells expressing haematopoietic genes such as SPN (despite being CD43 neg ) and MYB (Fig. 5b and Extended Data Fig. 5e,f).Similarly to Monocle3, two additional trajectory inference methods 40,41 depicted a unified developmental sequence with minimal deviations characterized by a FCGR2B + intermediate HEC state (Extended Data Fig. 5g-i).Collectively these analyses suggested that clusters 0, 1, 2, 11, 16 and 17 identified by scRNA-seq represent progressive developmental states of HECs within day 8 WNTd CD34 + CD43 neg CD184 neg CD73 neg cells.We next performed GSEA and ORA across this progression to dissect the unique features of FCGR2B + HECs and identify state-specific gene expression profiles proper of HECs with distinct characteristics.In particular, we observed that the gradual loss of endothelial identity of H19 + HECs begins with the downregulation of genes associated with extracellular organization, cell adhesion and cytoskeletal remodelling (Fig. 5c and Supplementary Tables 9-12).The progression to the FCGR2B + HEC cluster is associated with an enrichment of the expression of several ribosomal protein genes, consistent with the role of RUNX1 in regulating ribosome biogenesis 42 , a key process for the emergence of blood cells (Fig. 5c and Supplementary Tables 9-12).Next, concomitant with the upregulation of haematopoietic genes, HECs also display increased expression of genes associated with cell motility as well as cell cycle progression

Article
https://doi.org/10.1038/s41556-024-01403-0(Fig. 5c and Supplementary Tables 13-16).Indeed, using scRNA-seq data to infer the cell cycle state across HEC states, we observed that, during the progression from H19 + to FCGR2B + cluster (clusters 0, 1 and 2 to cluster 11), cells are mostly in G1 phase, while cells in the MYB + HEC clusters (clusters 16 and 17) are mostly in S/G2/M phase (Fig. 5d).Interestingly, the MYB + HEC clusters (that is, clusters 16 and 17) negatively correlated with the expression of genes associated with the activation of NOTCH signalling (Fig. 5c and Supplementary Tables 14-16).Since NOTCH signalling is an essential driver of stage-specific intra-embryonic emergence of haematopoietic cells 11 , we further analysed the expression trend according to pseudotime of the well-characterized NOTCH target genes in the DA, that is, HES1, HEY1 and HEY2 (Fig. 5e).The analysis revealed that HES1 expression peaks in cells belonging to cluster 11, marked by FCGR2B differential expression, while HEY1 and HEY2 expression peaks in cells that precede FCGR2B + cells in this pseudotime (Fig. 5e).This suggests that FCGR2B expression might identify a NOTCH-independent state of HECs.To test this hypothesis, we added the chemical γ-secretase inhibitor L-685,458 (γSi) to the HEC culture of day 8 WNTd CD32 + as well as CD32 neg cells, as some of the latter will generate CD32 + cells (Extended Data Fig. 3e).While CD32 neg cells gave rise to haematopoietic progenitors in a NOTCH-dependent manner, the chemical inhibition of NOTCH signalling did not impair the generation of CD45 + haematopoietic progenitors from CD32 + cells (Fig. 5f,g).Altogether these results show that CD32 expression defines the temporal NOTCH requirement within the HEC continuum and that HECs activate a cell division programme to give rise to haematopoietic progeny.
To further elucidate the role of CD32 in the ontogeny of HECs, we performed an in silico perturbation analysis of CD32 across the six identified HEC clusters using CellOracle 41 .By simulating a knockout of FCGR2B and observing the resultant effects on both direct and indirect gene targets, we identified a shift in the developmental trajectory of HEC differentiation, particularly noting a reversion in differentiation around cluster 11 (Extended Data Fig. 6a-c).This pattern suggests that a perturbation of CD32 signalling is likely to affect the differentiation process at these critical stages of HEC development.Therefore, we silenced FCGR2B expression in H1 hESCs (herein CD32 knock-down, KD) introducing a knockdown construct into the AAVS1 'safe harbour' locus to perform functional studies 43 (Extended Data Fig. 6d).This strategy effectively reduced the expression of CD32 without affecting the overall day 8 CD34 + cell output (Extended Data Fig. 6e,f).Upon differentiation, CD32 KD cells displayed a significantly reduced haematopoietic output compared with H1 wild-type cells (Fig. 5h and Extended Data Fig. 6g).This suggests that, under these conditions, in the absence of FCGR2B the efficiency of haematopoietic differentiation is severely reduced.

Stage-specific regulators of haematopoietic development
We then leveraged the specificity of CD32 expression on HECs to identify pathways driving HEC specification that can be harnessed to increase the haematopoietic output from hPSC cultures.ORA analysis of the transcriptomic profiles of the CD32 + and DLL4 + hPSC-derived populations described above revealed that CD32 + cells are positively associated with BMP signalling GO terms (Fig. 6a and Supplementary Table 17).This led us to hypothesize that the development of CD32 + HECs may require active BMP signalling.We therefore tested the effects of either the activation or the inhibition of BMP4 signalling during the HEC specification stage (day 3 to day 8, Extended Data Fig. 3a) in WNTd hPSC differentiations.While the addition of BMP4 resulted in around a twofold increase of the proportion of CD32 + cells at day 8, BMP signalling inhibition (BMPi) severely impaired their formation (Fig. 6b,d).We functionally validated the role of BMP signalling during HEC specification by testing the haematopoietic potential of day 8 CD144 + cells isolated from BMP-or BMPi-treated WNTd haematopoietic cultures, as it should reflect the variation of the proportion of CD32 + cells observed.As expected, CD144 + cells isolated from BMP-treated cultures gave rise to twofold more CD45 + cells and haematopoietic progenitors after HEC culture.On the other hand, BMP signalling inhibition almost abrogated the haematopoietic potential of day 8 CD144 + cells (Fig. 6c,e).
We next leveraged the hPSC-derived scRNA-seq dataset described above to identify pathways regulating RUNX1 + HECs progression to the blood fate.We focused on the progression to FCGR2B + state (that is, from clusters 0, 1 and 2 to cluster 11, Fig. 5b) given that CD32 marks a unique HEC stage.ORA revealed that this progression is associated with a downregulation of Rho/ROCK signalling (Fig. 6f and Supplementary Table 18).We then tested whether ROCK signalling inhibition during the HEC culture could increase the haematopoietic output from hPSCs.Indeed, ROCK signalling inhibition increased the proportion of CD45 + and clonogenic progenitors generated during HEC culture (Fig. 6g,h), probably by synchronizing or facilitating the progression of HECs towards the blood fate.Collectively, these results indicate that HEC specification requires stage-specific BMP signalling activation and Rho/ROCK signalling inhibition, highlighting the value of using CD32 to study HEC biology.

HEC CD32 expression is conserved across developmental programmes
Since CD32 is also expressed in YS endothelial cells where it enriches for a population with haematopoietic potential (Fig. 2d,f), we investigated whether CD32 is a conserved marker of HECs with robust haematopoietic potential across different haematopoietic programmes specified from hPSCs.For this, we used the hPSC differentiation protocol that includes a stage-specific inhibition of WNT signalling by IWP2.This leads to the emergence by day 8 of WNT-independent (WNTi) HOXA neg/low extra-embryonic-like HECs, whose gene expression and potential are similar to the YS-derived EMP haematopoietic programme 9,35,36,44 .

Discussion
The precise identification of human HECs will enable the characterization of this transient population and unveil what regulates their transition to blood, thus allowing us to determine their identity, which is still subject to debate.In this study, using transcriptomic analysis of haemogenic populations sorted from human embryos, we identified FCGR2B, which encodes for a CD32 isoform, as a marker whose expression is upregulated in HECs.CD32 expression can be used in combination with other endothelial markers to precisely isolate HECs with robust haematopoietic potential from both the human embryo and hPSC differentiating cultures.In fact, CD32 expression is more specific for hPSC-derived HECs than CD44, another marker known to be expressed in HECs as well as in arterial cells 13,20,21 .The use of a hPSC-based model has allowed us to capture and dissect in fine detail the HEC progression to the generation of blood cells.By providing new granularity to a rare process that in vivo occurs very rapidly, our study demonstrates that HECs can be found in different intermediate states, which display transcriptional profiles indicative of distinct cellular characteristics.The progression of HECs towards the haematopoietic fate begins with a gradual change in the expression of genes associated with the remodelling of the extracellular matrix, the loss of adhesion molecules and the re-organization of the cytoskeleton (Fig. 5c).This suggests that the transcriptional changes associated with the release of haematopoietic cells in the bloodstream occur before the actual morphological remodelling can be observed.In addition, this HEC progression culminates with the enhancement of ribosome biogenesis and translation in CD32 + committed HECs that appear to be irreversibly fated to haematopoiesis.Indeed, CD32 + HECs can generate haematopoietic progeny independently of NOTCH signalling, the major driver of this process (Fig. 5f).As such, our study defines the temporal requirement of NOTCH signalling to initiate the haematopoietic programme in intra-embryonic HECs, with CD32 expression demarcating pre-NOTCH versus post-NOTCH states during HEC progression to the blood fate.
The initiation of the morphological remodelling and the concomitant expression of haematopoietic markers are often identified as the moment of the haematopoietic specification of HECs 45 .However, our data suggest that the commitment of HECs to the blood fate is temporally distinct from, and becomes NOTCH-independent before, the full execution of the haematopoietic programme, which occurs at a different cell cycle state.In fact, while the fate decision coincides with a timely suppression of the cell cycle, the emergence of blood cells is associated with cell cycle re-entry (Fig. 5d).Given that the acquisition of an active cell cycle to generate lineage output is a hallmark of cell differentiation 46 , rather than developmental lineage transition 47,48 , our findings support a model in which blood cells emerge via the differentiation of haematopoietic-restricted HECs rather than an endothelial-to-haematopoietic transition.In this model (Fig. 6i), we propose that HECs are specified from mesodermal cells in a vascular endothelial growth factor (VEGF)-and BMP-dependent manner.HECs expressing RUNX1 isoforms driven by the P2 proximal promoter then enter the determination stage, which is a process driven by BMP and NOTCH signalling.During this process, which occurs in G1 phase, RUNX1 + HECs increase their translation and gradually downregulate Rho/ROCK signalling to become CD32 + .This marks the irreversible commitment of HECs that progressively upregulate the expression of haematopoietic genes while re-entering the cell cycle to upregulate RUNX1C and differentiate into blood cells in a NOTCH-independent manner.
Our studies suggest that CD32 plays a functional role in the development of the human haematopoietic lineage.More refined studies are needed to determine the exact timing of its requirement and how mechanistically it exerts its role.In addition, the fact that in immunodeficient humans HECs generate functional blood cells in the absence of circulating antibodies, the best-characterized ligands of CD32, suggests that CD32 could potentially functionally regulate blood development via the binding of alternative ligands.
Our findings indicate that CD32 is expressed in HECs harbouring multilineage haematopoietic potential isolated from different anatomic locations of the human embryo and in HECs derived from hPSC differentiations recapitulating distinct haematopoietic programmes (Figs.2f and 3g, and Extended Data Fig. 7c).These data suggest that HEC progression to a CD32 + stage before generating blood cells, is a conserved developmental process across ontogenies.However, we could not test whether CD32 is expressed on HSC-competent HECs, since culture conditions supporting human HSC specification from HECs, even from human embryonic explant cultures, have not been identified yet 49 .Indeed, additional challenging experiments are needed to formally prove a direct lineage relationship between CD32 + HECs and HSCs using stage-specific human embryos.As such, despite the functional results described, we cannot exclude that HECs can give rise to blood cells, potentially HSCs, independently of CD32 signalling.
In summary, this study demonstrates that expression of CD32 marks HECs fully committed to generate haematopoietic progeny and suggests that it could be used as a powerful tool to enrich haematopoietic precursors from a broad range of hPSC lines, including those for which current differentiation protocols into haematopoietic lineages are not optimal.Our findings will allow a deeper understanding of the specification of the HEC lineage, a central element of haematopoietic development, which will translate into optimized scaled-down, potentially more cost-effective, protocols to generate therapeutic blood products from hPSCs.

Ethics declaration
The use of human embryonic tissues described in this study is compliant with the International Society for Stem Cell Research guidelines.All human embryonic tissue samples used in this study were discarded material from elective terminations that were obtained once informed written consent to the use of samples in research was obtained from patients.The donated human embryonic tissues were anonymized and did not carry any personal identifiers.In all cases, the decision to terminate the pregnancy occurred before the decision to donate tissue.No payments were made to donors, and the donors knowingly and willingly consented to provide research materials without restrictions for research and for use without identifiers.Human embryonic tissues employed for RNA-seq, immunohistochemistry and immunofluorescence were obtained from voluntary abortions performed according to the guidelines and with the approval of the French National Ethics Committee.The study was approved by Ospedale San Raffaele Ethical Committee (TIGET-HPCT protocol) and by the Institutional Review Board of the French Institute of Medical Research and Health (number 21-854).Human embryonic tissues employed for ex vivo haematopoietic cultures were collected by the Human Developmental Biology Resource (HDBR; HDBR project number 200430), Newcastle University, Newcastle, United Kingdom, with approval from the Newcastle and North Tyneside NHS Health Authority Joint Ethics Committee (08/H0906/21 + 5).The HDBR is regulated by the UK Human Tissue Authority (HTA; https://www.hta.gov.uk/) and operates in accordance with the relevant HTA Codes of Practice.This use was also approved by Ospedale San Raffaele Ethical Committee (TIGET-HPCT protocol).No embryos were created nor cultured for research purposes.The use of hESCs was approved by the Ospedale San Raffaele Ethical Committee, included in the TIGET-HPCT protocol.

Human embryonic tissues
Human embryos were staged using anatomic criteria and the Carnegie classification.Samples employed for RNA-seq, immunohistochemistry and immunofluorescence were either used immediately as fresh tissues (ex vivo experiments and RNA-seq analysis) or fixed in phosphate-buffered saline (PBS) supplemented with 4% paraformaldehyde (Sigma-Aldrich), embedded in gelatin and stored at −80 °C (immunohistochemistry and immunofluorescence).Human embryonic tissues (CS12-CS13) analysed by RNA-seq were incubated in medium containing 0.23% w/v collagenase Type I (Worthington Biochemical Corporation, NC9482366) for 30 min at 37 °C, and the single-cell suspensions were filtered through a 70 μm cell strainer (BD Biosciences).

Immunohistochemistry and immunofluorescence
The techniques employed have been previously described 19 .Briefly, 5-μm sections were incubated first with primary antibodies overnight at 4 °C, then for 1 h at room temperature (RT) with biotinylated secondary antibodies and finally with fluorochrome-labelled (Bio-Legend) or peroxidase-labelled streptavidin (Beckman Coulter).
Peroxidase activity was revealed with 0.025% 3,3-diaminobenzidine (Sigma-Aldrich) in PBS containing 0.03% hydrogen peroxide.Low amounts of antigens (CD32 and ACE) were revealed by Tyramide signal amplification biotin or fluorescence amplification systems (Akoya, Biosciences).An isotype-matched negative control was performed for each immunostaining.When 3,3-diaminobenzidine was used on slides, they were counterstained with Gill's haematoxylin (Sigma-Aldrich), mounted in XAM neutral medium (BDH Laboratory Supplies), analysed and imaged using an Optiphot 2 microscope (Nikon).Immunofluorescence-stained sections were cover-slipped in Prolong Gold Antifade Mountant with DAPI (Thermo Fisher Scientific) and analysed with an Axio Imager M2 microscope coupled to a Hamamatsu's camera Orca Flash 4v3 using the ApoTome.2function (Zeiss) for optical sectioning.The antibodies employed are listed in Supplementary Table 19.

RNA-seq
Human embryo sorted cells were collected in 6 μl of PBS supplemented with 0.5 μl of Protector RNAse inhibitor (Roche, 3335399001) and conserved in −80 °C.Full-length coding DNA (cDNA) was generated using Clontech SMART-Seq v4 Ultra Low Input RNA kit for Sequencing (Takara Bio Europe, 634891) according to the manufacturer's instructions with 15 cycles of PCR for cDNA amplification by Seq-Amp polymerase.A total of 600 pg of pre-amplified cDNA was then used as input for Tn5 transposon tagmentation by the Nextera XT DNA Library Preparation Kit (Illumina, FC-131-1096) followed by 12 cycles of library amplification.After purification with Agencourt AMPure XP beads (Beckman Coulter, A63882), the size and concentration of libraries were assessed by capillary electrophoresis.Libraries were then sequenced with the Illumina HiSeq 4000 sequencing platform in the single-end mode and with a read length of 50 bp.For day 8 WNTd hPSC-derived haematopoietic cultures, total RNA from sorted CD34 + CD32 + CD43 neg CD184 neg CD73 neg DLL4 neg and CD34 + CD184 + CD73 + DLL4 + CD43 neg was purified using the ReliaPrep RNA Cell Miniprep System and RNA-seq libraries were generated using the Smart-seq2 method.One nanogram of RNA was retrotranscribed, and cDNA was PCR amplified (15 cycles) and purified with AMPure XP beads.Sequencing was performed on an Illumina NovaSeq6000 (single-end, 100 bp read length) following the manufacturer's instruction.
For both RNA-seq datasets, raw reads quality control was accomplished using the FastQC tool (http://www.bioinformatics.babraham.ac.uk/projects/fastqc) and read trimming was performed using the Trim Galore software (https://doi.org/10.5281/zenodo.5127899)to remove residual adapters and low-quality sequences.Trimmed reads were aligned against the human reference genome (GRCh38) using STAR 51 with standard parameters.Uniquely mapped reads were then assigned to genes using the featureCounts tool from the Subread package 52 , considering the GENCODE primary assembly v.34 gene transfer file as reference annotation for the genomic features.Gene count matrices were then processed by using the R/Bioconductor differential gene expression analysis packages DESeq2 (ref.53) applying the standard workflow.
For the human embryos' dataset, a paired analysis was set up modelling gene counts using the following design formula: ~donor + condition.Gene P values were corrected for multiple testing using false discovery rate (FDR).Genes with adjusted P values <0.05 were considered differentially expressed.
ORA and a GSEA were then computed considering the GO Biological Process (BP) terms from the C5 collection of the Molecular Signatures Database (MSigDB version 7.2) using the R/Bioconductor package clusterProfiler 54 (http://bioconductor.org/packages/ release/bioc/html/clusterProfiler.html,v 3.8.1).ORA was applied to the significantly DEGs, while GSEA was performed by pre-ranking genes according to fold change (FC) values.P values were corrected for multiple testing using FDR and enriched terms with an adjusted https://doi.org/10.1038/s41556-024-01403-0P value less than 0.05 were considered statistically significant.Volcano plots were generated using the R package ggplot2 (https://ggplot2.tidyverse.org)and have been used to display RNA-seq results plotting the statistical significance (adjusted P value) versus the magnitude of change (FC).Heatmaps were generated using the R package pheatmap (https://CRAN.R-project.org/package=pheatmap).Surface genes were extracted using the surfaceome database (http://wlab.ethz.ch/surfaceome/) 55 .

scRNA-seq
CD34 + CD43 neg CD73 neg CD184 neg cells were sorted at day 8 from WNTd hPSC-derived haematopoietic cultures that were treated at day 2 with 6 μM SB-431542 (Tocris, 1614) 56 .Libraries were prepared following the manufacturer's instructions using the Chromium platform (10x Genomics) with the 3′ gene expression (3′ GEX) V3 kit, using an input of ~10,000 cells.Briefly, Gel-Bead in Emulsions (GEMs) were generated on the sample chip in the Chromium controller.Barcoded cDNA was extracted from the GEMs by post GEM reverse transcription cleanup and amplified for 12 cycles.Amplified cDNA was fragmented and subjected to end-repair, poly A-tailing, adapter ligation and 10x-specific sample indexing following the manufacturer's protocol.cDNA libraries were sequenced in paired-end mode on a NovaSeq instrument (Illumina) targeting a depth of 50,000-100,000 reads per cell.
Sequencing reads were processed into gene count matrix by Cell Ranger (https://support.10xgenomics.com/single-cell-gene-expression/soft ware/pipelines/latest/ what-is-cell-ranger, v 4.0.0)from the Chromium Single Cell Software Suite by 10x Genomics.In detail, fastq files were generated using the Cell Ranger 'mkfastq' command with default parameters.Gene counts for each cell were quantified with the Cell Ranger 'count' command with default parameters.The human genome (GRCh38.p13) was used as the reference.The resultant gene expression matrix was imported into the R statistical environment (v 4.0.3)for further analyses.Cell filtering, data normalization and clustering were carried out using the R package Seurat 57 v 3.2.2.For each cell, the percentage of mitochondrial genes, number of total genes expressed and cell cycle scores (S and G1 phases) were calculated.Cells with a ratio of mitochondrial versus endogenous gene expression >0.2 were excluded as putative dying cells.Cells expressing <200 or >6,000 total genes were also discarded as putative poorly informative cells and multiplets, respectively.Cell cycle scores were calculated using the 'CellCycleScoring' function that assigns to each cell a score based on the expression of the S and G2/M phase markers and stores the S and G2/M scores in the metadata along with the predicted classification of the cell cycle state of each cell.Counts were normalized using Seurat function 'NormalizeData' with default parameters.Expression data were than scaled using the 'ScaleData' function, regressing on the number of unique molecular identifier, the percentage of mitochondrial gene expression and the difference between S and G2M scores.By using the most variable genes, dimensionality reduction was then performed with PCA by calculating 100 principal components (PCs) and selecting the top 55 PCs.Uniform manifold approximation and projection (UMAP) dimensionality reduction 58 was performed on the calculated PCs to obtain a two-dimensional representation for data visualization.Cell clusters were identified using the Louvain algorithm at resolution r = 0.6, implemented by the 'Find-Cluster' function of Seurat.To find the differentially expressed (marker) genes from each cluster, the 'FindAllMarkers' function (iteratively comparing one cluster against all the others) from the Seurat package was used with the following parameters: adjusted P values <0.05, average log FC >0.25, and percentage of cells with expression >0.1.A comprehensive manual annotation of the cell types was performed using the previously obtained markers list.DEGs between cells of cluster 11 against cells of clusters 0, 1 and 2 and clusters 16 and 17 were determined by the 'FindMarkers' function using the following parameters adjusted P values < 0.05, |average log FC| >0 and percentage of cells with expression >0.GSEA was then performed considering GO BP terms from the C5 collection of the Molecular Signatures Database (MsigDB v 7.2) using the R/Bioconductor package clusterProfiler 54 (http://bioconductor.org/packages/release/bioc/html/clusterProfiler.html,v 3.8.1).ORA was computed on the significantly DEGs considering GO BP terms from the C5 collection of the Molecular Signatures Database (MsigDB v 7.2) and the Reactome Pathways Database using the R/Bioconductor package 54 (v 3.8.1).P values were corrected for multiple testing using FDR and enriched terms with an adjusted P value less than 0.05 were considered statistically significant.A barplot was constructed using the R package ggplot2 (https://ggplot2.tidyverse.org).
scRNA-seq samples from the public dataset GSE162950 were retrieved and processed as described in ref. 38.The 'DotPlot' and the 'VlnPlot' functions from the Seurat R package were used to construct a scorecard highlighting the expression pattern of selected cells having RUNX1 + CDH5 + FCGR2B+ or RUNX1 CDH5 + FCGR2B− expression patterns.
Pseudotime trajectory was constructed using Monocle3 (https:// cole-trapnell-lab.github.io/monocle3/,v 0.2.3) 59,60 .Expression and feature data were extracted from the Seurat object, and a Monocle3 'cell_ data_set' object was constructed.The processed data were normalized followed by PCA analysis using the Monocle3 function 'preprocess_cds'.Dimensionality reduction was performed using the 'reduceDimension' function.Trajectory graph learning and pseudo-time measurement through reversed graph embedding were performed with 'learn_graph' function.Cells were ordered along the trajectory using the 'orderCells' method with default parameters.The 'plot_cells' function was used to generate the trajectory plots.
To corroborate the findings from Monocle3, a trajectory inference analysis was conducted using the dynverse workflow, a component of the R package dyno (v 0.1.2) 40.The Dynbenchmark utility, which offers a comprehensive framework for selecting the most suitable trajectory inference method according to the available experimental data, was utilized via the 'guidelines_shiny()' function.Following these guidelines, the trajectory inference analysis was performed using the partition-based graph abstraction (PAGA)-tree algorithm 61 .Input data for dyno, including gene expression matrices, dimensionality reduction coordinates, clustering information and cell metadata, were derived from Seurat output and processed using the 'wrap_expression()' function.The cell trajectory was subsequently calculated by dyno using the 'infer_trajectory()' function, employing the 'ti_paga_tree()' method.Trajectory paths and pseudotime values were visualized on UMAP coordinates through the 'plot_dimred()' function provided by dyno.
To investigate the role of the CD32 gene in HEC ontogeny, genetic knockouts were simulated using the CellOracle tool (v 0.12.0) 41.CellOracle integrates a gene regulatory network (GRN) with pseudotime analysis to predict shifts in cellular identities resulting from gene perturbations.This tool simulates alterations in gene expression due to perturbations and compares these changes with the cell's developmental trajectory within the GRN.This comparison allows for the estimation of transition probabilities between different cell states along the pseudotime axis.Following this, CellOracle generates a transition trajectory graph, illustrating the potential shifts in cellular identities after perturbation.This analysis was performed in a Python (version 3.8) environment, using Jupyter notebooks.scRNA-seq data, initially processed with Seurat, were converted to AnnData format using the anndata2ri tool (https://github.com/theislab/anndata2ri),ensuring content preservation for subsequent analysis.The CellOracle object construction utilized this data.Highly variable genes, critical for downstream analysis, were identified using the scanpy.pp.filter_genes_dispersion() function from scanpy 62 , specifying n_top_genes = 3,000.A preliminary GRN was constructed using the oracle.get_links()function within the Oracle() class, based on ligand-receptor interactions from the CellTalkDB database 63 .This base GRN was further refined by incorporating the CD32 gene and its interactors, as identified in https://doi.org/10.1038/s41556-024-01403-0 the STRING database (https://string-db.org/).Pseudotime analysis was conducted using the Pseudotime_calculator() class, employing the PAGA method from scanpy and integrating it into the CellOracle framework.This analysis culminated in the creation of a pseudotime gradient vector field with the Gradient_calculator() class from Cel-lOracle, depicting the normal developmental trajectory.Subsequently, in silico perturbation of CD32 expression and simulation of resultant cell identity shifts were performed using the simulate_shift() and esti-mate_transition_prob() functions from the Oracle class.To compare the effects of CD32 perturbation with normal development, the Ora-cle_development_module class was used to calculate perturbation scores by computing the inner product of the respective vector fields with the calculate_inner_product() function.
Single CD34 + CD43 neg CD184 neg CD73 neg DLL4 neg CD32 + /CD44 + cells were FACS-sorted directly onto a Matrigel-coated well of 96-well plate at day 8 of WNTd haematopoietic cultures.Cells were cultured as above.Haematopoietic and non-haematopoietic clones were evaluated by light microscopy and FACS analysis after 10-14 days of culture.

Cell staining, flow cytometry and cell sorting
Samples for FACS analysis or cell sorting were incubated with antibody mixes for 15-30 min at 4 °C.Dead cells were excluded using 7-aminoactinomycin D (7AAD) during staining.For the analysis of the T cell maturation, ATO aggregates were stained with Maleimide PromoFluor840 for dead cell exclusion.Cells were then incubated with antibody mixes diluted in Brilliant stain buffer (BD Biosciences, 563794) + PBS (Corning, 21-040-CM) + 2% FBS (HyClone, SH30066.03)+ 4% FcR blocking for 15 min at RT.After washing, cells were fixed in 1% paraformaldehyde and analysed.The antibodies employed are listed in Supplementary Table 19.Cells were sorted with FACSAria II with the FACSDiva software (BD Biosciences).Sorting gates were set using appropriate fluorescence minus one and single staining controls.FACS analysis was performed using FACS Canto with the FACSDiva software (BD Biosciences) or Cytoflex S or Cytoflex LX with Coulter CytExpert software (Beckman Coulter) for the acquisition and the FlowJo software (BD Biosciences) for the analysis.Where indicated, WNTd day 8 CD144 + cells were selected using magnetic bead-based separation with CD144 Micro-Beads (Miltenyi Biotec, 130-097-857) following the manufacturer's instructions.

Statistics and reproducibility
For all multivariate statistical analyses, analyses of variance (ANOVAs) were performed with the appropriate corrections for multiple comparisons.One-way ANOVA with Tukey's multiple comparison test was chosen for single metrics with more than two populations.The data distribution was not formally tested but was assumed to be normally distributed with equal variance.Sample size and replication were determined by historical controls 14 .For bivariate statistical analyses, Student's t-test was performed with the appropriate corrections (one tail, non-parametric).In general, biological replicates were excluded only if internal controls failed and technical replicates were not excluded.Experimental conditions were not randomized, but covariates were controlled by an equal distribution of sorted cells across controls and experimental conditions.Blinding of experimental conditions was not relevant as our studies do not require grading of the results.

Reporting summary
Further information on research design is available in the Nature Portfolio Reporting Summary linked to this article.
Positive regulation of mitotic cell cycle Mitotic nuclear division Mitotic cell cycle checkpoint Positive regulation of cell cycle phase transition Positive regulation of cell cycle G2 M transition Positive regulation of cell cycle Positive regulation of cell cycle process Regulation of endothelial cell migration Endothelial cell migration Regulation of epithelial cell migration Positive regulation of epithelial cell migration Tissue migration Amoeboidal type cell migration Leukocyte migration Extracellular matrix binding Leukocyte cell-cell adhesion Regulation of cell-cell adhesion Negative regulation of cell adhesionCell-

Fig. 1 |
Fig.1| Human embryonic ACE + endothelial cells express arterial and haemogenic markers.a, A transverse section of the AGM region of a 23 dpf (CS10, top, n = 2 independent) and a 27 dpf (CS12, bottom, n = 3 independent) human embryo, immunostained with ACE (left, in red), RUNX1 (middle, in green) and merge (right).Ao, aorta; NT, neural tube.Scale bars, 50 μm.b-e, An RNA-seq analysis of human embryonic populations isolated from four CS12-CS13 embryos, referred to as 'donor': E1, E2, E3 and E4.The ACE neg population is coloured in beige and ACE + population in light blue.PCA (b) of the top 500 DEGs within human embryonic populations.A heatmap of DEGs within human embryonic populations (c), where gene counts were corrected for donor and

Fig. 2 |
Fig. 2 | CD32 is expressed in AGM and YS HECs during human embryonic development.a, A heatmap of top ten differentially expressed surface genes within human embryonic ACE + (light blue) and ACE neg (beige) cells derived from four CS12-CS13 human embryos (E1, E2, E3 and E4).rlog gene expression values are shown in rows.Colouring indicates differential expression by upregulation (red) or downregulation (blue).b, CD34 (top) and CD32 (bottom) expression by immunohistochemistry of consecutive sections of the AGM of a 29 dpf (CS13) human embryo (n = 5 independent).Inset shows high magnification of haematopoietic clusters (black arrowhead) and surrounding endothelial cells (white arrowhead) immunostained by CD32.Ao, aorta.Scale bars, 50 μm and 25 μm (inset).c, Transverse consecutive sections of the AGM region of a 29 dpf

Fig. 4 |
Fig. 4 | Transcriptomic similarity between CD32 + cells and AGM-derived HECs.a, A heatmap showing a selection of DEGs in CD34 + CD184 + CD73 + DLL4 + CD43 neg (DLL4 + , in green) or CD34 + CD43 neg CD184 neg CD73 neg CD32 + (CD32 + , in orange) samples isolated at day 8 of WNTd hPSC-derived haematopoietic culture.H1 hESCs, n = 3 independent (#1 in light blue, #2 in purple and #3 in yellow).Scaled rlog gene expression values are shown in rows.The colouring indicates differential expression by upregulation (red) or downregulation (blue).b, UMAP of integrated publicly available single-cell datasets (accession codes GSE135202 and GSE162950) showing cell clusters of AGM (CS10-CS16) or YS (CS11).Cells clustered at resolution 1.2.c, Feature plots showing HOXA9 and HOXA10 expression across the clusters shown in b. d, A barplot displaying the transcript per million (TPM) values for HOXA9 and HOXA10 genes analysed from the RNAseq data of CD32 + cells as described in a, mean ± s.e.m. e, A heatmap visualizing the similarity scores between AGM samples from the single-cell data shown in b and annotated as HEC (red) or AEC (blue) in comparison with CD32 + and DLL4 + samples.Scale bar: Z scores of the relative Spearman coefficients.Each column is representative of a single embryonic cell scored across each population indicated by the row name.f, A scorecard dot plot showing landmark genes as reported in ref. 38.Differential expression was evaluated in CDH5 + RUNX1 + PTPRC neg cells that either express FCGR2B (FCGR2B > 0) or do not (FCGR2B = 0).