Hematopoietic stem and progenitor cells (HSPCs) balance the physiological demands of maintaining peripheral leukocytes, erythrocytes, and platelets, while maintaining a potent stem cell reserve. These characteristics have made HSPC transplantation the only curative option for treating many hematological disorders. The regenerative potential of hematopoietic stem cells (HSCs) reside in their ability to home to a supportive niche, allowing for both HSC self-renewal and reconstitution of the hematopoietic system. The bone marrow (BM) microenvironment regulates HSC quiescence, self-renewal, and differentiation. The BM niche is composed of a number of cell types, including Lepr+ cells, Nestin+ cells, and endothelial cells. Collectively, the HSC niche modulates HSC fate decisions through the expression of paracrine factors, including Cxcl12 and Kitl. The BM microenvironment also plays a critical role in the reestablishment of hematopoiesis following myeloablative injury. Vegfr2-mediated vascular repair is critical for hematopoietic reconstitution following chemotherapeutic and radiation-mediated insult, while BM granulocyte production of Tnfa supports the regeneration of sinusoidal endothelium and subsequent hematopoietic recovery. While the importance of an HSC-supportive microenvironment during hematopoietic homeostasis and during regenerative conditions is coming into focus, poorly defined BM niche cells have limited the precise mechanistic insights necessary to elucidate new regenerative factors and strategies. Current methodology used to examine cellular subsets within the BM microenvironment rely on immunophenotypic fractionation, localization, and genetic lineage tracing. Ambiguous BM niche cellular immunophenotypes and gene expression have limited cellular resolution and confounded the interpretation of cre-mediated genetic deletion models. Herein, we aim to resolve the identities of distinct BM endothelial cell (BMEC) subpopulations to ultimately develop genetic tools to elucidate the paracrine requirements of the HSC-supportive endothelial niche. To this end, we sort-purified murine BMECs (VECAD+CD31+CD45-TER119-) for single cell RNA sequencing (scRNA-Seq). scRNA-Seq revealed the emergence of distinct BM arteriole, sinusoidal, and transitional endothelial populations, with arteriole BMECs significantly enriched for Kitl and Cxcl12. To confirm an arteriole enrichment in Kitl and Cxcl12 expression, we performed scRNA-Seq transcriptional analysis of sort-purified BM cells from KitlGFP and Cxcl12DsRed reporter mice. KitlGFP BM cells identified a distinct arteriole, but not sinusoidal, BMEC population. Cxcl12DsRed BM cells identified both arteriole and sinusoidal BMEC cell populations, but confirmed an increase in Cxcl12 expression in arterioles. We next examined candidate genes to generate BMEC subset-specific inducible cre mice. Analysis revealed that Vegfr3 (Flt4) expression was specific to sinusoidal BMECs, while Bmx1 appeared enriched in arterioles. We utilized a previously described Vegfr3YFP transgenic reporter mouse and found sinusoidal BMEC restricted expression. We then generated an inducible Vegfr3creERT2 line that directs efficient recombination to sinusoidal endothelium, with no detectable off-target activity. We next examined a previously described Bmx1creERT2 mouse line by generating Bmx1creERT2;ROSA26tdTomato reporter mice. In contrast to a recent report, Bmx1creERT2 activity was not spatially confined to the BM arteriole niche, but also labeled additional niche components, making it an unsuitable for arteriole-specific deletion. Moreover, previously reported constitutive Eporcre mice used to delete Kitl in sinusoids displays detectable cre activity in both arteriole and erythrocyte populations. More refined genetic models will need to be generated to test current and future candidate factors in the BM niche. Using our transcriptional data set, we have generated and validated a new inducible Vegfr3creERT2 mouse line that displays sinusoidal-restricted expression in the BM. Arteriole-specific creERT2 lines are currently being evaluated. These models will be used to systematically evaluate novel candidate arteriole- and sinusoidal-specific hematopoietic paracrine factors identified in our transcriptional analysis.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.