HSCs are mesodermal in origin and derive from a subset of endothelial cells (ECs) known as the hemogenic endothelium (HE) in the major embryonic arteries, including vitelline and umbilical arteries and the aorta (Medvinsky et al., 2011). The specific requirement of a supportive niche in HSC specification has been most directly demonstrated in avian embryos by the use of quail-chick chimeras (Pardanaud and Dieterlen-Lievre, 1999), in mice via in vitro explant-reaggregates cultures (Matsuoka et al., 2001; Peeters et al., 2009; Souilhol et al., 2016a; Taoudi and Medvinsky, 2007; Ivanovs et al., 2014) and in Xenopus via transplantation (Turpen et al., 1997). Table 2 summarizes the methods that have been employed to identify hematopoietic niches during development. We also refer the reader to our previous review article focused on classic and novel approaches to study the bone marrow niche (Sanchez-Lanzas et al., 2022). HSC specification mainly concentrates to the ventral aspect of the dorsal aorta (Jaffredo et al., 2013) (Figure 2). The establishment of a dorsoventral polarity in the dorsal aorta is key in HE commitment (Yvernogeau et al., 2023). It has been shown that the subaortic mesenchyme contributes to this process and during HSC specification (Jaffredo et al., 2013; Ivanovs et al., 2014; Chandrakanthan et al., 2022; Damm and Clements, 2017; Gonzalez Galofre et al., 2024; Kapeni et al., 2022; Miladinovic et al., 2024; Richard et al., 2013; Sa da Bandeira et al., 2022; Crosse et al., 2023; Crosse et al., 2020). The subaortic mesenchyme derives from the splanchnopleural mesoderm, as shown by studies on avian embryos, and contains mesenchymal cells harboring various differentiation potentials including the ability to generate osteoblasts, adipocytes and ECs (Jaffredo et al., 2013; Richard et al., 2013; Mendes et al., 2005; Durand et al., 2006; Pardanaud et al., 1996; Jaffredo et al., 2010). Cells with other embryonic origin also migrate to the subaortic mesenchyme and have a role on HSC specification (Figure 2). Particularly, migratory ectoderm-derived neural crest cells associate with the ventral aspect of the dorsal aorta before HSC specification in zebrafish embryos (Damm and Clements, 2017). Their migration requires platelet-derived growth factor receptor α (PDGFRα) signaling and blocking this migration leads to lack of HSC specification (Damm and Clements, 2017). Additionally, catecholamines produced by the sympathetic nervous system neurons (derived from a fraction neural crest cells) are required to maintain HSCs in the mouse embryo after HSC specification (Damm and Clements, 2017; Kapeni et al., 2022; Fitch et al., 2012). While in chick embryos somite-derived ECs migrate to the aorta to replace HE cells post-EHT (Pouget et al., 2006), in zebrafish dermomyotome-derived ECs were recently shown to travel to the dorsal aorta to support EHT from neighboring HE cells. It would be interesting to determine the timing and relevance of this migration in mammalian embryos (Sahai-Hernandez et al., 2023).

Within the subaortic mesenchyme, mesodermal-derived PDGFRα⁺ cells (Chandrakanthan et al., 2022) and NG2⁺ PDGFRβ⁺ cells (overlapping with NG2⁺RUNX1^LOWCD146⁺ cells) (Gonzalez Galofre et al., 2024; Sa da Bandeira et al., 2022) have been shown recently to support HSC specification, are proximal to the hemogenic endothelium and required for optimal specification (Clements and Khoury, 2024; Fadlullah et al., 2022). The EHT supportive ability of mesodermal-derived PDGFRα⁺ stromal cells seems to rely on PDGFRα, BMP, WNT and NOTCH signaling pathways (Chandrakanthan et al., 2022). Interestingly, mesodermal-derived PDGFRα⁺ stromal cells are present in the AGM during E10.5-E11.5, i.e., during EHT, and by E13.5 are replaced by neural crest-derived PDGFRα⁺ stromal cells (Chandrakanthan et al., 2022). In contrast to mesodermal-derived PDGFRα⁺, neural crest-derived PDGFRα⁺ stromal cells do not support EHT, as shown in reaggregate co-cultures (Chandrakanthan et al., 2022). As suggested by Chandrakanthan et al., this opens up the intriguing possibility that the replacement of mesodermal-derived with neural crest-derived PDGFRα⁺ stromal cells constitutes a regulatory mechanism selected to actively control HSC specification confining it to a very defined developmental window (Chandrakanthan et al., 2022). Interestingly, in zebrafish, the PDGFβ/PDGFRβ axis mediates HIF1α-regulated signaling axis leading to IL-6/IL-6R activation and controls embryonic HSPC production (Lim et al., 2017).

The generation of stromal clones from the surrounding aortic tissues uncovered the presence of critical factors as bone morphogenic protein-4 (BMP4) (Oostendorp et al., 2002; Oostendorp et al., 2005; Ohneda et al., 2000; Renstrom et al., 2009; Istvanffy et al., 2011), highly expressed in the aortic ventral mesenchyme of avian, mouse, zebrafish and human embryos (Pimanda et al., 2007; Marshall et al., 2000) and active in transplantable HSCs in mouse embryos at embryonic day of development (E) 11 (Crisan et al., 2015). Moreover, BMP signaling contributes to AGM polarity in zebrafish (Wilkinson et al., 2009). The use of explant cultures derived from the aorta-gonad-mesonephros region (AGM), developed by the laboratory of Alexander Medvinsky (Taoudi and Medvinsky, 2007; Taoudi et al., 2008; McGarvey et al., 2017), allowed to show that BMP4 is required in HSC specification as HSC potential was devoid in AGM explants treated with gremlin (a BMP antagonist) (Durand et al., 2007). Analysis of gene regulatory networks supported that the BMP signaling pathway and the SCL transcriptional network regulate RUNX1 activity (Pimanda et al., 2007). Importantly, the subaortic mesenchyme triggers RUNX1 expression on aortic ECs. RUNX1 marks the hemogenic endothelium and is critical for HSC specification and dispensable in later hematopoietic stages (Chen et al., 2009; North et al., 2002; Lancrin et al., 2009). Furthermore, GATA3 binds to enhancer elements in the Runx1 locus in the HSC microenvironment, suggesting a role for RUNX1 also in the stromal cells that regulate EHT (Fitch et al., 2020).

Interestingly, BMP signaling must be downregulated following EHT to allow further HSC maturation and mediated at least in part by NOGGIN in mice (Souilhol et al., 2016a) and FGF signaling in zebrafish (Pouget et al., 2014). Additional work in avian, zebrafish and mouse embryos unveiled that the Hedgehog (Hh) signaling pathway regulated by ventral tissues is key in HSC specification (Pardanaud and Dieterlen-Lievre, 1999; Peeters et al., 2009; Wilkinson et al., 2009; Gering and Patient, 2005) while dorsal tissues exert an opposing effect, dorsoventrally polarizing the aorta (Jaffredo et al., 2013). The subaortic mesenchyme would be the source for ventral BMP4 (Marshall et al., 2000; Wilkinson et al., 2009; Durand et al., 2007), while Sonic Hedgehog (Shh) would be produced in the developing gut (Peeters et al., 2009) and both responsible for HSC specification in the ventral aspect of the dorsal aorta. Interestingly, in zebrafish embryos, the Hh pathway inhibits the hematopoietic program later during development in the dorsal aspect of the aorta, which highlights the fine-regulation of HSC specification (Wilkinson et al., 2009).

Furthermore, WNT16-dependent somite patterning and NOTCH signaling activation by DeltaC and DeltaD contributes to HSC specification (Clements et al., 2011). The intermediate mechanistic steps are still unknown (Clements and Khoury, 2024; Clements et al., 2011). The urogenital ridges in the AGM in mice also promote HSC specification although the exact involved mechanisms remain unclear (Clements and Khoury, 2024; Souilhol et al., 2016a).

Before the fusion of the paired aortas, all the aortic ECs present at that stage derive from the splanchnopleural mesoderm (Jaffredo et al., 2013). Later during development there is a replacement of the dorsal ECs with somitic mesoderm derived ECs (Pardanaud et al., 1996; Pouget et al., 2006). Particularly, in seminal experiments, Pardanaud and Dieterlen-Lievre showed in avian embryos that dorsal ECs derive from the somitic mesoderm, while ventral ECs derive from the lateral mesoderm ECs, which contributes to the dorsoventral asymmetry (Pardanaud et al., 1996). In zebrafish and mice, the EC replacement precedes HSC specification (Chandrakanthan et al., 2022; Sahai-Hernandez et al., 2023; Nguyen et al., 2014) contributing to dorsoventral polarity and influencing HSC specification.

The hemogenic endothelium (HE) undergoes an EHT to render the first HSCs defined by their ability to engraft and repopulate the hematopoietic system of a lethally irradiated recipient. The first functional HSC is detected at E11 in mouse embryos and EHT can be visualized as c-Kit⁺ clusters budding into the lumen of the aorta forming intra-aortic clusters, IACs (Medvinsky et al., 2011) (Figure 2). IACs emerge from the ventral aspect of the dorsal aorta (Medvinsky et al., 2011; Medvinsky and Dzierzak, 1996). The lack of IACs in Runx1 ^−/− mouse embryos (Chen et al., 2009; North et al., 1999; Yokomizo et al., 2001) and the block in EHT observed in zebrafish embryos following runx1 inhibition by morpholino (Kissa and Herbomel, 2010) demonstrated a key role of Runx1 during EHT. NOTCH signaling is also required in EHT as shown via morpholino knockdown in zebrafish embryos (Bertrand et al., 2010). All components in the NOTCH signaling pathway including Notch ligands (Jag1 and Jag2), Notch1 receptor and NOTCH transcriptional targets (including Hes1 and Gata2) are expressed in the ventral aspect of the aorta (Robert-Moreno et al., 2005; Robert-Moreno et al., 2008; Hadland et al., 2022). GATA2 is critical in HSC emergence and regulates Runx1 expression (Nottingham et al., 2007), directly linking NOTCH activation with EHT. Furthermore, enforced Gata2 and Runx1 expression partially rescues in vitro the hematopoiesis defects driven by Notch1 and Jagged1 deficiency (Robert-Moreno et al., 2008; Nakagawa et al., 2006). Importantly, HSC specification is tightly connected with the arterial endothelial identity and suppression of NOTCH signaling via CoupTFII deficiency leads to the acquisition of arterial characteristics by the venous endothelium and emergence of venous hematopoietic clusters (You et al., 2005). In addition, JAG1 was shown to be required in embryonic hematopoiesis both in zebrafish and mice (Robert-Moreno et al., 2008). Similar to BMP signaling, NOTCH signaling needs to be downregulated to allow further HSC maturation (Richard et al., 2013; Souilhol et al., 2016b; Porcheri et al., 2020; Lizama et al., 2015; Ditadi et al., 2015; Kobayashi et al., 2014; Gama-Norton et al., 2015; Sacilotto et al., 2013; Thambyrajah and Bigas, 2022). Interestingly, the downregulation of NOTCH signaling may be driven by JAG1/NOTCH1 cis interaction (Ghersi et al., 2023).

During the recent years the role of sterile inflammatory signals on HSC specification have gained much attention (Hou et al., 2020; Espin-Palazon et al., 2014; Li et al., 2014; Mariani et al., 2019). Endothelial and myeloid cells (such as macrophages and primitive neutrophils) produced cytokines (including IFNα, IFNγ, NFκβ and TNFα) and affect HSC specification. Some of these cytokines (e.g., NFκβ and TNFα) converge on modulating NOTCH signaling in HSPCs in mice and zebrafish (Clements and Khoury, 2024; He et al., 2015) and are required in HSC specification. Additionally, myeloid populations contribute with matrix metalloproteases (Mmp2 and 9) that facilitate EHT and the migration of HSCs out from the dorsal aorta, as shown in zebrafish (Theodore et al., 2017). Interestingly, YS-derived macrophages, whose migration to the aorta depends on the chemokine receptor CX3CR1, constitute the largest hematopoietic cell population in the AGM and interact with IACs (Mariani et al., 2019). Notably, macrophage depletion in E10.5 AGM explant cultures blocks HSC specification in vitro (Mariani et al., 2019), supporting their role as part of the AGM HSC-specification niche.

Although it cannot be defined as a specific niche effect, the presence of blood circulation induces shear stress on the hemogenic endothelium, which activates the nitric oxide (NO) signaling pathway and plays an important role in EHT (Niranjan et al., 1995; Adamo et al., 2009; North et al., 2009). Blocking NO signaling pathway in NO synthase3 deficient mice and by the administration of L-NAME in vivo results in the lack of IACs (Adamo et al., 2009; North et al., 2009; Murayama et al., 2006). The absence of Runx1⁺ cells in zebrafish with no blood flow can be rescued via the administration of an NO donor (S-nitroso-N-acetylpenicillamine) further demonstrating that shear stress works through NO signaling pathway (North et al., 2009; Murayama et al., 2006). Moreover, compound BF170 hydrochloride has been shown to promote HSPC induction in vitro in zebrafish blastomere cell cultures and from mouse embryoid bodies by activating the NO signaling pathway (Liu et al., 2024).

Synchronously to the FBM, the fetal spleen starts to receive HSCs from the FL at E15.5 as shown by repopulating activity detected at E15.5 which increases until E17.5 and remains detectable over the first couple of weeks post-birth (Christensen et al., 2004; Morita et al., 2011; Ikuta and Weissman, 1992; Wolber et al., 2002). The specific role of these HSC reservoir in the spleen is unclear but splenectomy in neonatal mouse pups does not result in a reduction of HSC numbers in the perinatal BM, indicating that the spleen is not an intermediate migration location for HSCs before moving to the BM (Bertrand et al., 2006). Furthermore, organ culture systems of fetal spleens are not capable of sustaining HSC repopulating activity following in vitro culture (for 4 days) while FL can (Bertrand et al., 2006). Additionally, spleen-derived stromal cell lines support myeloid differentiation but not HSC repopulating activity indirectly indicating that the fetal spleen niche is less supportive than the FL niche (Bertrand et al., 2006).

The adult BM has been widely investigated over the last two decades. A large body of studies have shown that HSCs reside mostly in perivascular niches, both arteriolar and sinusoidal, during homeostasis in the adult BM (Sanchez-Lanzas et al., 2022; Morrison and Scadden, 2014; Lucas, 2017). Adult HSC niche components include sympathetic nerves, osteoblasts, stromal cells such perivascular stromal cells and sinusoidal endothelium, and hematopoietic cells including macrophages, megakaryocytes, T-regulatory cells and neutrophils (Sanchez-Lanzas et al., 2022; Casanova-Acebes et al., 2013; Nakamura-Ishizu et al., 2015; Fujisaki et al., 2011). Among stromal cells, leptin receptor LepR⁺ reticular cells and vascular ECs produce and secrete SCF which activates the c-KIT tyrosine kinase receptor in HSCs, key in regulating the self-renewal and maintenance of HSCs (Pinho and Frenette, 2019; Crane et al., 2017; Ramasamy et al., 2016; Mendez-Ferrer et al., 2020; Batsivari et al., 2020; Chen Q. et al., 2019; Liu et al., 2021; Ding et al., 2012; Comazzetto et al., 2019). Additionally, LepR⁺ cells are a source for stromal derived factor-1 (SDF1), also known as CXCL12, a chemokine critically required for homing, trafficking and maintenance of HSCs by signaling through the G-protein coupled receptor CXCR4 in HSCs (Greenbaum et al., 2013; Sugiyama et al., 2006; Lo et al., 2009). Other relevant signals and cytokines produced in the adult perivascular BM microenvironment include NOTCH ligands, Interleukin-7 (IL7), vascular endothelial growth factor (VEGF) and Angiopoietin-1 (AGPT1) (Fang et al., 2020; Tikhonova et al., 2019; Kusumbe et al., 2016; Poulos et al., 2013; Cordeiro Gomes et al., 2016; Zhou et al., 2015).

The FBM has been much less examined. Importantly, recent studies underline the presence of substantial differences among the adult and FBM both on the cellular components and at the molecular level (Hall et al., 2022; Langen et al., 2017) which have major functional implications on the behavior of the HSPCs.

Around birth, the downregulation of several niche factors from FL stellate cells coincides with enhanced migration of HSCs to the BM (Lee et al., 2021). Intriguingly, there is a burst in the number of FL-HSPCs at birth, which declines by P2 in the mouse embryo (Hall et al., 2022). By P14, HSPCs are basically absent in the liver (Hall et al., 2022). Simultaneously, it is only near birth, by E19, when FBM HSPCs become fully functional and exhibit a molecular identity close to adult BM HSCs (Hall et al., 2022). This correlates with a shift in the relative frequencies and numbers of the various HSPCs and the development of a supportive BM niche, as revealed by changes in the ability of mesenchymal stromal cell derived from P0 BM to sustain LT-HSCs expansion (Hall et al., 2022), as mentioned above. However, even though the neonatal BM starts to contain niche cells expressing factors that support HSPCs, these factors are different from those present in the adult BM, emphasizing critical differences among perinatal and adult BM(185) (Figure 6). Particularly, although Cxcl12⁺ cells are present in the perinatal BM, they produce lower CXCL12 levels than the adult BM Cxcl12⁺ cells (Omatsu et al., 2014). Likewise adult-like CXCL12-abundant reticular (CAR) cells are absent from the perinatal BM (Hall et al., 2022). As aforementioned, a subtype of osteochondral progenitors supports perinatal HSPC functions by producing IGF1/2 (Hall et al., 2022) (Figure 6). It would be interesting to understand the molecular mechanisms that drive and defines these changes in the postnatal bone marrow.

It is worth highlighting that at birth mammalian newborns are exposed to major environmental changes including significant hormonal fluctuations during labor and exposure to microbes and higher oxygen levels when they abandon sterile conditions in the uterus (Hall et al., 2022). Importantly, all these factors are known to impact hematopoiesis. Particularly, ambient oxygen levels cause HSC differentiation in vitro (Mantel et al., 2015). Furthermore, the composition of the gut microbiota plays a role in hematopoiesis (Dominguez-Bello et al., 2010; Josefsdottir et al., 2017), thus it is likely that the changes in the microbiome of the newborn shapes the HSC-niche. Moreover, as noted by Hall and colleagues (Hall et al., 2022), prostaglandins are upregulated during birth and they have been shown to modulate HSC expansion (Ratajczak and Muglia, 2008; North et al., 2007; Porter et al., 2013). Hence, it is very conceivable that they also influence HSC biology and the BM niche at this stage.

As FL-HSCs migrate to the FBM, HSC numbers increase until they reach a plateau around 3–4 weeks post-birth in the mouse. Importantly, HSCs are highly proliferative during the first 3 weeks post-birth in the neonatal BM before they undergo a quiescence shift that characterizes the adult state (Bowie et al., 2006). By 4 weeks, considering that the total number of BM nucleated cells is ∼2 × 10⁸ in a mouse and that HSC frequency is 1/10,000 per nucleated BM cells (Kiel et al., 2005; Trevisan et al., 1996; Sudo et al., 2000; Sieburg et al., 2006; Szilvassy et al., 1990; Boggs et al., 1982), a mouse harbors a total of ∼20,000 HSCs. The cellular mechanisms driving this increase in the pool of HSCs are incompletely understood and relevant to regenerative medicine. The specific weight that maturation, symmetric versus asymmetric cell division, differentiation and apoptosis have in this process requires further investigation (Ganuza et al., 2020).

Lineage tracing studies tracking the global clonal complexity dynamics of hematopoietic progenitors exposed that the number of HSCs with lifelong contribution to the HSC pool expands by ∼2 fold from birth to P21 when the quiescence shift happens in mice (Ganuza et al., 2022a; Bowie et al., 2006) (Figure 1).

Similar dynamics have been detected in human based studies. Analysis of the effect of age in X-chromosome inactivation patterns in healthy women (Catlin et al., 2011) and in telomere lengths in granulocytes and lymphocytes in healthy individuals (Werner et al., 2015; Busque et al., 1996; Busque et al., 2012), showed that HSC numbers mostly expand from birth until adolescence and divide once every 40 weeks (Catlin et al., 2011) with a progressive bias towards the accumulation of HSPCs with myeloid potential with age (Werner et al., 2015). Whole genome sequencing of clones of expanded human BM-HSPCs and reconstruction of phylogenetic trees based on the analysis of the natural accumulation of somatic mutations allowed to investigate the clonal complexity origins and lineage relationships in the hematopoiesis system in humans (Lee-Six et al., 2018; Osorio et al., 2018). These studies revealed that between 50,000 and 2,00,000 clones participate in adult steady-state hematopoiesis in humans (Lee-Six et al., 2018), compatible with clonal dynamics observed in mice (Ganuza et al., 2019). Furthermore, these analyses indicated that most of the expansion in the HSC pool occurs during childhood and adolescence (Lee-Six et al., 2018). In agreement with a quiescence shift at that stage, HSC numbers would stabilize in the adulthood (Lee-Six et al., 2018). These studies also predict the presence of HSPC clones with persistent myeloid and B-lymphocyte in life, while dynamics on T-lymphoid potential were not as conclusive (Lee-Six et al., 2018; Osorio et al., 2018).

The expansion of the HSC pool may be allowed by “the development of the BM vasculature and adult niches” (Gao et al., 2018; Arai and Suda, 2007; Wilson et al., 2008) (Figure 7). In this regard, scRNA sequencing revealed that the frequencies of LepR⁺ stromal cells and ECs (arteriolar and sinusoidal) increase during the first weeks post-birth (analyzed by Kara et al. perinatally at P4, P14 and at 8 weeks post-birth) (Kara et al., 2023). These changes are accompanied by variations in the HSC-supportive ability of various populations (Kara et al., 2023). Functionally, confirming a role of Nestin⁺ cells and CXCL12 in the early postnatal BM niche, deletion of Cxcl12 in Nestin-CRE-ERT Cxcl12 floxed mice at P7 depletes HSCs (Isern et al., 2014), while Cxcl12 and Scf deletion from Nestin⁺ cells during adulthood does not perturb HSC numbers (Ding et al., 2012; Ding and Morrison, 2013). Likewise, Cxcl12 deletion from NG2⁺ cells at P14 depletes the HSC pool while its deletion during adulthood does not impact HSCs (Acar et al., 2015; Asada et al., 2017). Interestingly, Scf deletion from perinatal Nestin⁺ cells does not carry any defect (Kara et al., 2023). In the postnatal BM the patterns of expression of key adult BM factors are similar to those found in adult BM (Tikhonova et al., 2019; Baryawno et al., 2019; Baccin et al., 2020; Zhong et al., 2020; Matsushita et al., 2020). Particularly, perinatal BM LepR⁺ cells express the highest levels of Cxcl12, and both perinatal LepR⁺ and ECs exhibit the highest Scf levels, while Nestin⁺ and NG2⁺ cells do not constitute a main source of SCF (Kara et al., 2023). Interestingly, most ECs express Nestin in the perinatal BM (Mizoguchi et al., 2014; Ono et al., 2014) which becomes restricted to periarteriolar ECs and other stromal cells in the adult (Asada et al., 2017; Kunisaki et al., 2013). Notably, perinatal LepR⁺ cells-produced SCF promotes myelopoiesis and erythropoiesis but they are not required in HSC maintenance as shown by inducible conditional mouse models (Kara et al., 2023). Meanwhile, the use of Tie2-cre; Scf-Ex7 ^fl/fl mice, which enables to delete the transmembrane domain of SCF from ECs (Buono et al., 2016), showed that membrane-bound SCF in ECs is required to maintain the HSC pool in the perinatal BM (Kara et al., 2023). Still, this does not exclude that soluble SCF-produced by ECs plays also a role (Kara et al., 2023). Among adult BM-ECs, SCF is mostly expressed by arteriolar ECs (Tikhonova et al., 2019; Liu et al., 2022; Xu et al., 2018; Baryawno et al., 2019), whereas at P4, sinusoidal ECs also produce it (Kara et al., 2023). Confocal imaging of α-Catulin ^GFP/+ mice showed that, like in the adult BM, HSCs concentrate mostly in sinusoidal blood vessels in the perinatal BM (P4 and P14) (Kara et al., 2023). A small fraction of HSCs also locates to transition zone vessels, which exhibit “intermediate properties between sinusoids and arterioles” (Kara et al., 2023) (Figure 7). This may be reminiscent of the ongoing migration of HSCs from their arterial location in the FBM to sinusoids during adulthood (Liu et al., 2022).

Molecularly, important metabolic, transcriptional and epigenetic differences have been reported among FL-HSC an adult-BM HSCs which are likely linked to their proliferative stage and transition into quiescence (Manesia et al., 2015; Jassinskaja et al., 2017; Roy et al., 2021; Manesia et al., 2017). FL-HSCs contain more mitochondria, express higher levels of genes implicated in oxidative phosphorylation (OxPhos) and citric acid cycle (TCA), consume higher oxygen and produce increased ROS levels than BM HSCs (Agrawal et al., 2024; Manesia et al., 2015). Interestingly, adult HSCs display a larger collection of proteins involved in the protection against ROS-induced protein oxidation that help to preserve HSC stemness in the long-term (Jassinskaja et al., 2017). Inflammatory signaling is also upregulated in adult HSCs (Roy et al., 2021). Remarkably, Sox17 is expressed in fetal and perinatal HSCs until ∼4 weeks post-birth concomitant with the quiescence shift, and conditional knockout of Sox17 leads to the loss of fetal HSCs but does not impact adult HSCs when deleted later in development. This underscores the critical role of Sox17 in fetal HSCs (Kim et al., 2007).

Epigenomic analyses revealed that even though chromosomal compartments and topologically associating domains (TADs) are broadly conserved between fetal and adult HSCs, there is still an increased compartmentalization in the chromosomal compartments and more dynamic chromatin interactions within TADs in adult HSCs (Chen C. et al., 2019). This affects thousands of interactions among gene promoters and enhancers. These changes are driven by particular transcription factors, including TCF3 and MAFB in fetal HSCs and NR4A1 and GATA3 in adult HSCs (Chen C. et al., 2019).

Overall, these investigations emphasized major molecular and functional differences among fetal and adult HSCs which are highly determined and regulated by developmentally distinct HSPC niches.