Hematopoietic stem and progenitor cells (HSPCs) that originate from the hemogenic endothelium in the dorsal aorta finally home and engraft within the adult niche, the fetal bone marrow. Within the bone marrow, HSPCs are retained in a complex microenvironment surrounded by niche cells such as megakaryocytes, peripheral nerves, endothelial and mesenchymal stromal cells. While it is known that niche cells secrete hematopoietic factors that are essential for HSPCs to self-renew and repopulate the blood lineages, the ultrastructure of this niche is not well defined as current imaging technology does not allow direct visualization of the fetal bone marrow niche. Zebrafish share a similar hematopoietic ontogeny to mammals, and because the embryos are transparent, intrinsic HSPC interactions with the niche can be directly visualized.

To precisely locate rare HSPCs within the large dense kidney marrow, the presumptive adult niche, we genetically tagged endogenous HSPCs to track them live using lightsheet microscopy, followed by high-resolution serial block-face scanning electron microscopy (SBEM) (XY = 10.8 nm/pixel, Z = 70 nm/pixel). Using this technique, we could visually track single mCherry+ HSPCs, then confirm their exact location in the SBEM dataset with high contrast APEX2 peroxidase label. We found HSPC clusters within vessel lumens, as well as single HSPC in a novel perivascular niche. In this perivascular site, a single HSPC simultaneously contacted one mesenchymal stromal cell, multiple endothelial cells, a glial-like cell, and other hematopoietic cells. After extensive tracing of the glial-like cells within the 3D SBEM data, we identified that these cells extended as a long chain and formed contact with HSPCs. Our search for candidate structures in the kidney marrow region suggested that the glial-like cells could be dopamine beta-hydroxylase positive (dbh+) cells. Through fluorescence imaging, we observed cytoplasmic projections extending from dbh+ cells into Runx+ HSPC clusters. To confirm the identity of these glial-like cells, we performed fluorescence imaging of dbh+ transgenic larvae followed by SBEM. By correlating the two datasets, we identified dbh+ glial-like cell in contact with a single HSPC in the niche. Furthermore, treatment with a neurotoxin, 6-hydroxydopamine, led to a significant reduction in HSPC numbers within the niche highlighting the biological significance of dbh+ cells.

In summary, through this approach, we could identify a previously unknown cell type within the zebrafish adult hematopoietic niche that is pivotal in maintaining the HSPC pool. Our finding validates the importance of having multiple niche cells that concurrently regulate HSPCs within their niche. Importantly, our approach can be applied to characterize the ultrastructure of rare cells, such as other stem cells, and their niche by using multiple imaging modalities and can also lead to the identification of previously uncharacterized cell types. Further, we can now identify novel intercellular structures that form between an unperturbed HSPC in its endogenous perivascular niche.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.