Abstract
Hematopoietic stem cells (HSCs) self-renew and differentiate into all blood types in response to various demands through life. HSC functions are tightly and finely tuned by a specialized microenvironment called "niche" in the bone marrow (BM). Using Nestin-GFP transgenic mice, we have identified Nestin-GFP+ perivascular stromal cells exhibiting a mesenchymal stem/progenitor cell activity as niche cells. Furthermore, we found two types of Nestin-GFP+ cells expressing different surface markers, Nerve/glial antigen 2 (NG2) and Leptin receptor (Lepr) that are associated with arterioles and sinusoid, respectively, in the BM (Kunisaki et al. Nature, 2013). Both arteriolar and sinusoidal niche cells have been reported to show high gene expression of cytokines essential for HSC maintenance such as CXCL12 and stem cell factor (SCF), however, it remains unknown how the distinct niche cells differentially regulate HSC functions. To investigate the mechanisms, we utilized genetic mouse models, in which CXCL12 or SCF can be deleted in specific cell types. CXCL12 deletion in sinusoidal niche cells by using Lepr-cre/Cxcl12fl/− mice mobilized HSCs and lineage− Sca-1+ c-kit+ (LSK) progenitors into spleen (HSC, CT: Control/DL: Deleted: 760±165 / 2193±557 / spleen, n=6, p<0.05) and blood (LSK, CT/DL: 177±36 / 668±156 / mL blood, n=5, p<0.05), but had no effect on HSC numbers in the BM (CT/DL: 1435±101 / 1194±75 / femur, n=6, p=0.085), which is consistent with a previous report (Ding and Morrison, Nature, 2013). Furthermore, assessments of endogenous HSC localization using whole-mount 3D imaging technology revealed that the deletion of CXCL12 in Lepr+ niche cells had no impact on HSC location (KS-test: Two-sample Kolmogorov-Smirnov test, p=0.9981). By contrast, deletion of CXCL12 in NG2-cre derived cells, which recombines efficiently in the entire Nestin-GFP+ non-endothelial stromal fraction including both peri-arteriolar and peri-sinusoidal cells (96.9±1.3%), and overlapping with Lepr+ cells (88.5±1.6%) and CXCL12-abundant reticular cells (90.7±1.4%), led to a robust reduction of HSC numbers in the BM (CT/DL: 1487±87 / 179±40 / femur, n=10, p<0.0001) with HSC and progenitor mobilization into spleen (HSC, CT/DL: 705±262 / 3550±540 / spleen, n=6-8, p<0.01) and blood (LSK, CT/DL: 494±178 / 5357±896 / mL blood, n=5-7, p<0.01). In addition, deletion of CXCL12 in NG2-cre targeted cells led to HSC displacement away from arterioles (KS-test: Two-sample Kolmogorov-Smirnov test; p=0.001). To examine further a role of CXCL12 produced by NG2+ arteriolar niches on HSC maintenance, we generated tamoxifen-inducible NG2-creERTM/Cxcl12fl/− mice. Deletion of CXCL12 postnatally in NG2+ arteriolar niche cells significantly reduced the number of HSCs in the BM (CT/DL: 1617±160 / 960±95 / femur, n=10-13, p=0.0013), which was confirmed functionally by a competitive repopulation assay. Moreover, 3D imaging revealed that HSCs were located further away from arterioles in NG2-creERTM/Cxcl12fl/− marrow (KS-test: p<0.0001), suggesting a role for arteriolar niches in CXCL12-mediated HSC maintenance. As niche cells synthesize several factors, we evaluated the contribution of arteriolar niches in SCF synthesis, a cytokine shown to be critical for HSC maintenance. As expected, deletion of SCF in NG2-cre targeted cells led to a significant reduction of HSC numbers in the BM (CT/DL: 606±85 / 96±23 / femur, n=5-7, p<0.0001). To further evaluate functions of SCF produced by distinct vascular niches, we also compared these mice with deletions using Lepr-cre or tamoxifen-inducible NG2-creERTM mice. We found that deletion of SCF in Lepr-cre targeted cells showed a significant reduction of HSC numbers in the BM (CT/DL: 690±84 / 220±83 / femur, n=3-4, p<0.0118), consistent with previous studies (Ding et al., Nature, 2012), whereas there was no significant change observed in NG2-creERTM/SCFfl/− mice, suggesting that Lepr+ vascular niches rather than NG2+ arteriolar niches are the most important source of SCF in the BM. These results highlight distinct contributions of perivascular cells primarily located in separate vascular niches, arteriolar and sinusoidal, in HSC maintenance and mobilization.
No relevant conflicts of interest to declare.
Author notes
Asterisk with author names denotes non-ASH members.