Osteoclasts promote the formation of hematopoietic stem cell niches in the bone marrow

Abstract

Formation of the hematopoietic stem cell (HSC) niche in bone marrow (BM) is tightly associated with endochondral ossification, but little is known about the mechanisms involved. We used the oc/oc mouse, a mouse model with impaired endochondral ossification caused by a loss of osteoclast (OCL) activity, to investigate the role of osteoblasts (OBLs) and OCLs in the HSC niche formation. The absence of OCL activity resulted in a defective HSC niche associated with an increased proportion of mesenchymal progenitors but reduced osteoblastic differentiation, leading to impaired HSC homing to the BM. Restoration of OCL activity reversed the defect in HSC niche formation. Our data demonstrate that OBLs are required for establishing HSC niches and that osteoblastic development is induced by OCLs. These findings broaden our knowledge of the HSC niche formation, which is critical for understanding normal and pathological hematopoiesis.

Figure 1.

The HSC pool is dramatically reduced in the oc/oc BM. (A) Flow cytometry analysis of LSK cells in BM of 17-d-old oc/oc and control littermates gated on Lin^neg cells. Gating and percentage of the LSK cells are indicated. (B) The percentage and the absolute number of BM LSK cells from the hind legs was determined and presented as the mean ± SD from six mice per group. *, P < 0.01. (C) Analysis of CD45 expression within the Lin^negSca1⁺c-kit^neg BM population. Data are representative of those obtained for six mice in each group and three independent experiments.

Figure 2.

Phenotype of the mesenchymal cells from the BM of oc/oc mice. (A) Flow cytometry analysis of BM Ter119^negCD45^neg mesenchymal cells. Bar graph shows the mean ± SD of the percentage of mesenchymal Ter119^negCD45^neg cells obtained from 6 17-d-old oc/oc and control littermates per group and is representative of 2 independent experiments. *, P < 0.01. (B) Flow cytometry analysis of CD45 and Sca1 expression in Ter119^neg BM cells from +/+ and oc/oc mice. Bar graph shows represents the mean ± SD of the percentage of cell populations obtained for 6 mice per group and are representative of 2 independent experiments. *, P < 0.01. (C) Flow cytometry analysis of LSK (top) and Lin^negSca1⁺c-kit^neg (bottom) cells in the BM of 15-d-old +/+ mice treated with PBS or ZA. Data are representative of those obtained for three mice in each group in two independent experiments.

Figure 3.

Impaired homing of hematopoietic progenitors in oc/oc mice. (A) Real-time RT-PCR analysis on CD45^neg cells sorted from the BM of oc/oc and wild-type mice. Ct values were normalized to the 36B4 RNA. Differences were calculated with the 2^−ΔCt method and expressed as the percentage relative to the values obtained for the +/+ cells. Results are presented as the mean ± SD of triplicates from 10 oc/oc and 5 +/+ mice. *, P < 0.01. (B) In vitro migration assay of LSK cells sorted from the BM of actin-GFP (3–4-wk-old) mice in response to CD45^neg mesenchymal cells sorted from the BM of oc/oc and +/+ littermate (17-d-old) mice. Results are presented as the mean ± SD of triplicates of cells pooled from three oc/oc and three wild-type mice and are representative of two independent experiments. *, P < 0.05. (C) In vivo homing analysis of LSK sorted from the BM of actin-GFP mice and injected (2.5 × 10⁵ cells) into newborn oc/oc and control mice. Analysis of GFP⁺ cells was performed 18 h after cell transfer. Percentages of GFP⁺ cells are indicated. The data are representative of three oc/oc and control mice and two independent experiments. (D) Flow cytometry analysis of LSK cells in the spleen and the blood of 17-d-old +/+ and oc/oc mice. Cells were gated on Lin^neg cells. Percentage of LSK cells is indicated. Bar graphs show the mean ± SD of LSK cell percentage obtained for six mice per group and are representative of two independent experiments. *, P < 0.01.

Figure 4.

Reduced osteoblastic commitment in the BM of oc/oc mice. (A) Flow cytometry analysis showing CD44, PDGFRα, CD49e expression on Ter119^negCD45^negSca1⁺ cells and CD51 expression on Ter119^negCD45^neg cells from the BM of oc/oc and wild-type mice. Percentages of the populations are indicated, and the data are representative of three mice per group and two independent experiments. (B) Real-time RT-PCR analysis of Runx2, Alp, Osteocalcin, and Bsp expression in CD45^neg cells sorted from the BM of oc/oc and wild-type mice. Ct values were normalized to the 36B4 RNA. Differences were calculated with the 2^−ΔCt method and expressed as the percentage relative to the values obtained for the wild-type cells. Results are presented as the mean ± SD of triplicates from 10 oc/oc and 5 wild-type mice. *, P < 0.01. (C) Flow cytometry analysis of F4/80⁺ and Ly6G⁺ macrophages populations in the BM of oc/oc and +/+ mice. Cells were gated on CD11b⁺ cells. Percentages of the populations are indicated and are representative of those obtained from three mice per group and two independent experiments. (D) Flow cytometry analysis of F4/80⁺CD115⁺Gr1^negCD169⁺ macrophages in the BM of oc/oc and +/+ mice. Cells were gated on total BM cells (top), on Gr1^neg cells (middle) and on Gr1^neg F4/80⁺CD115⁺SSC^int/low cells. Percentages of the populations are indicated and are representative of those obtained from three mice per group and two independent experiments.

Figure 5.

Restoration of OCL activity rescues LSK homing in the BM of oc/oc mice. Sorted CD45⁺Sca1⁺ hematopoietic progenitors from actin-GFP (3–4-wk-old) mice were transferred into newborn oc/oc and wild-type mice. Analyses were performed at day 45 for the wild-type and treated oc/oc mice (oc/oc + CD45⁺Sca1⁺), and at day 17 for the untreated oc/oc mice because of their short life span. (A) Typical FACS profile of sorted GFP⁺CD45⁺Sca1⁺ cells before transfer into irradiated recipient mice. (B) Radiological analysis of the femur of treated and control mice, representative of four mice per group in two independent experiments. (C) Flow cytometry analysis of CD45 and GFP expression in the BM of the wild-type and treated oc/oc mice. Percentage of the cells in each quadrant is indicated. Data are representative four mice in each group in two independent experiments. (D) Flow cytometry analysis of LSK cells in the BM of the treated +/+ and oc/oc mice and control untreated oc/oc mice. Cells were gated on Lin^neg cells. Percentages of the populations are indicated. Data are representative four mice per group in two independent experiments. (E) Bar graph shows the mean ± SD of the percentage of LSK cells in the BM and the spleen obtained for four mice per group in two independent experiments. *, P < 0.01 (+/+ versus control oc/oc mice); **, P < 0.01 (treated oc/oc versus control oc/oc mice). (F) Mean ± SD of the percentage of Sca1⁺c-kit^neg cells in Lin^neg cells from the BM and the spleen of four mice per group in two independent experiments. *, P < 0.01. (G) Flow cytometry analysis on BM LSK cells of 17-d-old PBS- or DC-treated oc/oc mice gated on lin^neg cells (top) and on CD45 expression gated on BM Lin^negSca1⁺c-kit^neg cells (bottom). Data are representative of those obtained for three mice per group in one experiment. (H) Flow cytometry analysis of BM Ter119^negCD45^neg mesenchymal cells. Data are representative of those obtained for three mice per group in one experiment. (I) Analysis of GFP expression in LSK cells from the wild-type and oc/oc mice treated with CD45⁺Sca1⁺GFP⁺ cells. Cells were gated on BM LSK cells. Bar graph shows the mean ± SD of the percentage of GFP⁺LSK cells obtained for four mice per group in two independent experiments. *, P < 0.05. (J) Flow cytometry analysis of the main hematopoietic lineages (B220 for B cells, CD11b for monocytes, and CD3 for T cells) in the BM of oc/oc mice treated with CD45⁺Sca1⁺GFP⁺ cells. Percentage of cells in each quadrant is indicated. Data are representative of four treated oc/oc mice in two independent experiments.

Figure 6.

Restoration of OCL activity rescues the mesenchymal phenotype in the BM of oc/oc mice. (A) Flow cytometry analysis of CD45 expression in Ter119^neg BM cells from control and oc/oc mice treated with CD45⁺Sca1⁺GFP⁺ cells presented in Fig. 4. Cells were gated on Ter119^neg cells. Data are representative of four mice per group in two independent experiments. (B) Mean ± SD of the percentage of Ter119^negCD45^neg Sca1⁺ cells in the BM of four mice per group and representative of two independent experiments. *, P < 0.01. (C) Histological analysis of the tibia after Van Gieson and Alcian blue staining; type 1 collagen (pink) and cartilage (blue). Bars, 200 µm. Data are representative of two mice per group in two independent experiments. (D) RT-PCR analysis of CD45^neg cells from treated and control mice. Ct values were normalized to the 36B4 RNA. Differences were calculated with the 2^−ΔCt method and expressed as percentage relative to the values obtained for the wild-type cells. Results are presented as the mean ± SD of triplicates from four mice in each group and are representative of two independent experiments. *, P < 0.01 (treated oc/oc versus control oc/oc mice). (E–H) Histological analysis on the BM of +/+ mice (E), oc/oc mice (F and G), and treated oc/oc mice (H) using purple TRAP staining for OCLs (arrows) and blue ALP staining for osteoblastic cells. Data are representative of three mice per group in two independent experiments. Bars, 50 µm.