The Aorta-Gonad-Mesonephros niche shapes the functions of yolk sac-derived macrophages involved in hematopoietic stem and progenitor cell generation ex vivo

Background

Hematopoietic stem cell transplantation (HSCT) has become a viable treatment option in patients suffering from haematological disorders and malignancies¹. However, transfusion of healthy hematopoietic stem cells (HSCs) from self (autologous), an identical twin (syngeneic) or a healthy donor (allogeneic) comes with its own unique challenges, hindering its full utilisation potential. Persistent effort therefore has been made to develop and generate HSCs from induced pluripotent stem cells (iPSCs). But fully functioning, truly engraftable HSCs capable of maintaining multi-lineage blood cells have not been generated ex vivo^2,3. One reason is that the development of embryonic HSCs takes place within a highly specialised microenvironment that remains to be recreated accurately in vitro.

During ontogenetic development in mice, the first “primitive” hematopoietic cells are created in the yolk sac (YS) as early as E7⁴. The first “definitive” HSCs are found in the aorta-gonad-mesonephros (AGM) region around E10.5⁴. They are generated from endothelial cells with hemogenic potential to form blood cells (Endothelial-to-hematopoietic transition; EHT)⁴. Previous studies have shown that macrophages originating from YS populate the AGM during the Erythroid-Myeloid Progenitor wave (EMP). Whether these macrophages are intrinsically programmed to support HSC formation, or instead acquire these properties once in the presence of the AGM niche, remains unclear.

In this preprint, Belmonte et al. investigate how AGM macrophages contribute to HSC generation and whether the embryonic niche itself instructs the macrophages to become supportive of HSC generation. To investigate the origin, maturation and function of AGM macrophages, the authors combined flow cytometry, lineage-tracking, transcriptomic analysis and co-culture assays in mouse models.

Key findings

Refining the identity of AGM macrophages

The authors first asked whether the CD206^– myeloid population was composed of only macrophages. Using cell sorting in MacGreen mouse embryos, they identified a subpopulation of CD206^– cells expressing low levels of CD11b (mature myeloid marker) and F4/80 (murine macrophage marker), while retaining ckit expression (stem cell marker). This indicated that the population comprised immature myeloid progenitors as well. Almost all CD206⁺ cells had lost ckit expression and displayed a mature macrophage phenotype. These findings prompted the authors to incorporate CD11b, F4/80 and ckit into their gating strategy to distinguish myeloid progenitors from the mature CD206⁺ macrophages.

AGM macrophages originate in the yolk sac

To ascertain whether macrophages found in AGM are generated there or not, the authors performed lineage-tracing using Cdh5—CreERT2; Roas26-LSL-tdTomato mouse embryos. They labelled the YS hemogenic endothelial-derived progenitors at E7.5, resulting in tdTomato expression in 96-100% of CD206⁺ and CD206^– AGM populations at E10.5. At earlier stages, similar but more variable labelling was observed. After normalising against YS endothelial-myeloid progenitor labelling, it was observed that labelling frequencies at all stages were comparable. This provided evidence supporting a common YS origin, and provided no evidence for de novo macrophage production in the AGM at this developmental stage.

F4/80 regulates the AGM microenvironment supporting haematopoiesis

The authors next investigated whether F4/80 has a functional role beyond identifying and marking mature macrophages. For this, they created an F4/80 knockout mouse model. They observed a reduction in the endothelial population alongside an increase in the differentiated hematopoietic progenitor population in the AGM only. The colony-forming potential was assessed by a CFU-C assay, and F4/80- deficient AGM cells produced significantly more than the wild type used as the control. These observations were not evident in the YS, making the role of F4/80 niche-specific as opposed to ontogeny-related.

Interestingly, this suggests that interactions between AGM macrophages and developing endothelial and hematopoietic cells may help regulate the balance between these progenitor populations during embryonic development.

AGM macrophages acquire specialised functions from their surrounding niche

Using co-culture experiments, the authors compared the CD206⁺ macrophages isolated from the AGM and YS. When AGM endothelial cells were cultured with AGM macrophages, they generated hematopoietic cells, whereas the YS populations did not exhibit the same effect. AGM CD206⁺ macrophages promoted the generation of hematopoietic progenitors more effectively as opposed to their YS counterparts, suggesting that the developmental role was environment-specific and not simply because of their origin.

Transcriptomic analysis highlighted differences between the two macrophage populations, reinforcing the idea that the two populations were molecularly different. Compared with the YS macrophages, AGM macrophages highly expressed inflammatory mediators, such as Ccl2, Ccl9, Ccl12, Pf4, and Tnfα. This shows that the macrophages within the AGM manifest a more distinct, pro-inflammatory phenotype. Taken together, these findings support a model in which YS-generated macrophages enter the AGM and are eventually shaped by their environment and demonstrate varying functional properties.

Conclusions & Future Directions

This study helps delve deeper into the fundamental concepts of developmental biology: cellular origin may not always determine cellular function. Although the AGM macrophages originate in the YS, their interactions and functions are strongly influenced by the AGM niche (Figure 1). Whether similar niche-driven organ development takes place elsewhere is yet to be determined.  This work opens several exciting opportunities for regenerative medicine. The candidate signalling molecules identified, including Mmp2, Ccl2, Cxcl16 and Nrep, provide an attractive starting point for improving the current iPSC-derived HSC development and differentiation protocols.

Figure 1: Same origin, different function: Yolk-sac derived macrophages acquire different functions within the AGM niche. Created by author in Microsoft PowerPoint. Adapted from the preprint.

Why I liked this preprint

Directed cell migration during embryogenesis is a well-identified process. However, how the tissues around influence the function and identity of the incoming cells is still an open question. I particularly enjoyed how the authors differentiated between cellular origin and cellular function as part of this work. Rather than assuming that macrophages in the AGM would perform the same function as the ones in YS because they share a developmental origin, the authors show that the AGM niche can shape their molecular and functional properties.

From a developmental biology point of view, I found this particularly interesting. This perspective highlights the niche as an active participant in cellular fate and function, rather than just the location where the macrophages migrate to. The finding that the yolk-sac derived macrophages acquire specialised properties within the AGM raises the possibility that similar niche- and environment-dependent functional programming may exist during the development of other tissues. The study not just provides insight into haematopoiesis, but also offers a broader perspective on how these interactions can generate tissue-specific functions. The potential application of these findings to improve HSPC generation from iPSCs is also exciting.

Questions for the authors

1. Do you think the niche-specific programming in AGM is reversible, depending on where they migrate to later?
2. Given that the macrophages with a common developmental origin can acquire distinct functions in the AGM, do you think the macrophages involved in gonads or kidney formation will exhibit different functional properties as well?
3. Given the inflammatory genes being upregulated in the AGM macrophages, do you think these inflammatory signals directly support HSC development?

References

1. Aljagthmi AA, Abdel-Aziz AK. Hematopoietic stem cells: Understanding the mechanisms to unleash the therapeutic potential of hematopoietic stem cell transplantation. Stem Cell Res Ther [Internet]. 2025 Dec 1;16(1):60. Available from: https://pmc.ncbi.nlm.nih.gov/articles/PMC11809095/
2. Demirci S, Leonard A, Tisdale JF. Hematopoietic stem cells from pluripotent stem cells: Clinical potential, challenges, and future perspectives. Stem Cells Transl Med [Internet]. 2020 Dec 1;9(12):1549. Available from: https://pmc.ncbi.nlm.nih.gov/articles/PMC7695636/
3. Seiler K, Tsuneto M, Melchers F. Experimental Limitations Using Reprogrammed Cells for Hematopoietic Differentiation. J Biomed Biotechnol [Internet]. 2011 ;2011:895086. Available from: https://pmc.ncbi.nlm.nih.gov/articles/PMC3237023/
4. Kauts ML, Vink CS, Dzierzak E. Hematopoietic (stem) cell development — how divergent are the roads taken? Febs Lett [Internet]. 2016 Nov 1 ;590(22):3975. Available from: https://pmc.ncbi.nlm.nih.gov/articles/PMC5125883/