Skip to main page content
U.S. flag

An official website of the United States government

Dot gov

The .gov means it’s official.
Federal government websites often end in .gov or .mil. Before sharing sensitive information, make sure you’re on a federal government site.

Https

The site is secure.
The https:// ensures that you are connecting to the official website and that any information you provide is encrypted and transmitted securely.

Access keys NCBI Homepage MyNCBI Homepage Main Content Main Navigation
. 2013 Jan 23;32(2):219-30.
doi: 10.1038/emboj.2012.308. Epub 2012 Nov 27.

Identification of a clonally expanding haematopoietic compartment in bone marrow

Affiliations

Identification of a clonally expanding haematopoietic compartment in bone marrow

Lin Wang et al. EMBO J. .

Abstract

In mammals, postnatal haematopoiesis occurs in the bone marrow (BM) and involves specialized microenvironments controlling haematopoietic stem cell (HSC) behaviour and, in particular, stem cell dormancy and self-renewal. While these processes have been linked to a number of different stromal cell types and signalling pathways, it is currently unclear whether BM has a homogenous architecture devoid of structural and functional partitions. Here, we show with genetic labelling techniques, high-resolution imaging and functional experiments in mice that the periphery of the adult BM cavity harbours previously unrecognized compartments with distinct properties. These units, which we have termed hemospheres, were composed of endothelial, haematopoietic and mesenchymal cells, were enriched in CD150+ CD48- putative HSCs, and enabled rapid haematopoietic cell proliferation and clonal expansion. Inducible gene targeting of the receptor tyrosine kinase VEGFR2 in endothelial cells disrupted hemospheres and, concomitantly, reduced the number of CD150+ CD48- cells. Our results identify a previously unrecognized, vessel-associated BM compartment with a specific localization and properties distinct from the marrow cavity.

PubMed Disclaimer

Conflict of interest statement

The authors declare that they have no conflict of interest.

Figures

Figure 1
Figure 1
Gene targeting in the BM vasculature. (A) Maximum intensity projection of confocal images showing sinusoidal vessels in the femoral bone marrow cavity of a 3-month-old Cdh5(PAC)-CreERT2 x ROSA26-mT/mG mouse. Cre-induced mG signal marks endomucin-negative arterioles (arrows) as well as virtually all endomucin+ sinusoidal capillaries (arrowheads). (B) Confocal planes showing the association of mT+ perivascular cells (arrows in left image, red) with vessels in direct proximity of the growth plate (left) but not in the sinusoidal vasculature (arrowheads) within the marrow cavity of 3-month-old Cdh5(PAC)-CreERT2 x ROSA26-mT/mG mice. Image on the left shows an individual confocal plane of the inset in Figure 2A. Arrow in right image indicates an arteriole. SECs (green) and cell nuclei (Hoechst, blue) are labelled. Ch, chondrocytes.
Figure 2
Figure 2
Morphological features of hemospheres. (A) Maximum intensity projection of the metaphyseal region near the growth plate (gp) in a 3-month-old Cdh5(PAC)-CreERT2 × ROSA26-mT/mG mouse. ECs (mG, green), non-endothelial cells (mT, red) and cell nuclei (blue) are labelled. Right panel, inset at higher magnification. Vessels covered by mT+ cells (arrows) and trabecular bone marrow (bm) are indicated. (BD) Visualization of hemospheres in mT/mG mice. Individual confocal planes (right) and projection of Z-stack (left) are shown. Chondrocytes (ch) and bone marrow (bm) are indicated. (B) Smallest structures show the separation of the mG+ endothelial and mT+ perivascular layers (elongated nuclei) with enclosed putative haematopoietic cell (round nucleus, arrow). (C), More CD45+ (cyan) haematopoietic cells were found in the enlarged space between ECs and mT+ cells in bigger hemospheres (arrow), while mG+ and mT+ cells remained associated in the vessel outside this structure (arrowhead). The central capillary (arrow) was dilated in large hemospheres in comparison to the adjacent vessel (arrowhead). (D) In the largest hemospheres, the number of enclosed CD45+ cells was increased further. (E) Electron micrographs of hemospheres with central lumenized endothelium (asterisk), surrounding haematopoietic cells, peripheral osteoblasts (ob) and osteoclasts (oc), and enclosing bone (bn).
Figure 3
Figure 3
Enrichment of CD150+ CD48− cells in hemospheres. (A) 3D reconstruction of hemospheres in thick (100 μm) sections from a 6-week-old Cdh5(PAC)-CreERT2 x ROSA26-mT/mG femur. All channels (left) or only mG+ SECs (green, arrowhead in top panels) together with CD150 (cyan) and CD48 (magenta) immunofluorescence (right) are shown. CD150+ CD48− putative HSCs (arrows) and megakaryocytes (mk) are indicated. (B) Frequency of CD150+ CD48− cells (number per area) in hemospheres containing a total of 1, 2, 4, 8 or more of such cells, as indicated. Note that the CD150+ CD48− population was most concentrated in smaller hemospheres containing 2–4 of such cells. Equivalently sized areas (black bars) in the BM cavity hold only 1 or, in rare cases, 2 CD150+ CD48− cells. Error bars, s.e.m. (C) CD150+ CD48− cells inside hemospheres (HS) were predominantly associated with the abluminal surface of vessels. BM, bone marrow. Error bars, s.e.m. (D) Confocal section showing presence of CD150+ CD48− cells on the abluminal surface of a sinusoidal vessel inside hemosphere (lumen marked by asterisks). Mk, megakaryocyte. Right panel, higher magnification of inset. (E) Proposed model for the initiation and organization of hemospheres around distal sinusoidal vessels. SEC (green), perivascular mesenchymal cells (red), CD150+ CD48− cells (light blue) and other haematopoietic cells (purple and dark blue) are indicated.
Figure 4
Figure 4
Distinct properties of hemospheres. (A) Rapid EdU incorporation in haematopoietic cells in hemospheres (arrows) at the periphery of the bone marrow. Cells in the BM cavity (bm) are not labelled under the same conditions. Right, higher magnification of inset in left panel. (B) Quantitation of cells labelled by rapid EdU incorporation inside hemospheres (HS) or bone marrow (BM). Error bars, s.e.m. (C) Intrafemorally injected, mT+ haematopoietic cells (red) are found preferentially in hemospheres in proximity of SECs at day 5 (+5d) after transplantation (endomucin, green). Right image is higher magnification of inset. Nuclei, Hoechst (blue). (D) Distribution of transplanted BM cells at 5 days after transplantation in HS in comparison to BM. Error bars, s.e.m. (E) Intrafemorally injected haematopoietic cells (red) colonize the bone marrow cavity (arrows) at 10 days (+10d) after transplantation. Right image shows red fluorescence only. Nuclei, Hoechst (blue); SECs, endomucin (green).
Figure 5
Figure 5
Clonal haematopoietic cell expansion in hemospheres. (A) Genetic lineage tracing with Vav1-Cre and R26R-Confetti transgenics in spleen and BM of 2-week-old mice (top panels), and in 6-week-old vertebrae (centre) or femur (bottom). Indicating clonal expansion, cell clusters of one colour are dominant in hemospheres. Centre right image is higher magnification of inset on the left. (B) Dominance of uniformly coloured cells (1st) inside hemospheres compared to the other, less prevalent colours (2nd, 3rd, 4th). Error bars, s.e.m.
Figure 6
Figure 6
Morphological changes after loss of VEGFR2 activity. (A) Morphology of metaphysis and trabecular BM (bm) in vehicle (control) or SU5416-treated 6-week-old mice, as visualized with the indicated markers. (B) Reduction of distal vessel density and length, and capillary calibre in proximity of the growth plate of 8-week-old SU5416-treated animals. Error bars, s.e.m. (C) Morphological changes in the metaphysis of Flk1iΔEC mutants compared to control littermates (14-week-old mice), visualized with the indicated markers. Dotted line marks edge of trabecular BM. Small panels show mG (ECs, green) and Gr-1 (myeloid cells, cyan) immunosignals of the larger image above. (D) Quantitation of distal vessel density, length and capillary calibre in the Flk1iΔEC metaphysis at 14 weeks. Error bars, s.e.m.
Figure 7
Figure 7
Targeting of VEGFR2 disrupts hemospheres. (A) Distal sinusoidal vessels are found in direct proximity of chondrocytes (ch) in the control metaphysis. The equivalent region is devoid of SECs and contains only a few, collapsed hemospheres (arrow) in Flk1iΔEC mutants. Small panels show mG (ECs, green) and lineage-committed haematopoietic cells (Lin cocktail, cyan) immunosignals of the larger image above. (B) Quantitation of collapsed hemospheres in the Flk1iΔEC metaphysis and secondary ossifications centres (SOC). Error bars, s.e.m. (C) Quantitation of CD45+ haematopoietic cells per hemosphere in the secondary ossification centre of adult control and Flk1iΔEC mice, as indicated. Error bars, s.e.m. (D) Hemospheres in the SOC of control animals are filled with CD31+ endothelial (green/yellow) and CD45+ (cyan) haematopoietic cells. In contrast, Flk1iΔEC hemospheres appear and contain only very few CD45+ cells (arrows). Red signal, phalloidin (F-actin). Three corners in the left panel are filled in dark grey to cover regions without image data, which are the result of image rotation.
Figure 8
Figure 8
Altered putative HSC numbers after loss of VEGFR2 activity. (A) Analysis of LSK (Lin−, Sca-1+, cKit+) CD150+ Flk2− putative HSCs by multi-parametric flow cytometry (Supplementary Figure 12A) in the femoral BM of 24-week-old Flk1iΔEC mutants and control littermates under steady-state conditions. Error bars, s.e.m. (B) Quantitation of lineage-committed cells isolated from 24-week-old Flk1iΔEC mutants and control littermates. Mutants showed increased Ter-119+ erythroid cells and reduced CD11b+ macrophages, while the number of B220+ B cells, CD3e+ T cells and Gr-1+ granulocytes/neutrophils/macrophages was not altered compared to littermates. (C) Flow cytometric analysis of LSK CD150+ Flk2− cells in the BM of control and SU5416-treated 8-week-old C57BL/6 mice (Supplementary Figure 12C). Significant activation and mobilization of putative HSCs by CTX and G-CSF treatment was no longer obtained after SU5416 treatment. Statistics, one-way ANOVA and Holm-Sidak pairwise multiple group comparison. (D) Quantitation of LSK CD150+ Flk2− putative HSCs after CTX+G-CSF mobilization in the femoral BM of 14-week-old control and Flk1iΔEC mice. Error bars, s.e.m.

Comment in

References

    1. Armulik A, Abramsson A, Betsholtz C (2005) Endothelial/pericyte interactions. Circ Res 97: 512–523 - PubMed
    1. Becker AJ, McCulloch EA, Till JE (1963) Cytological demonstration of the clonal nature of spleen colonies derived from transplanted mouse marrow cells. Nature 197: 452–454 - PubMed
    1. Bianco P (2011) Bone and the hematopoietic niche: a tale of two stem cells. Blood 117: 5281–5288 - PubMed
    1. Blazsek I, Liu XH, Anjo A, Quittet P, Comisso M, Kim-Triana B, Misset JL (1995) The hematon, a morphogenetic functional complex in mammalian bone marrow, involves erythroblastic islands and granulocytic cobblestones. Exp Hematol 23: 309–319 - PubMed
    1. Bryder D, Rossi DJ, Weissman IL (2006) Hematopoietic stem cells: the paradigmatic tissue-specific stem cell. Am J Pathol 169: 338–346 - PMC - PubMed

Publication types

MeSH terms