Characterization of spontaneous bone marrow recovery after sublethal total body irradiation: importance of the osteoblastic/adipocytic balance

Abstract

Many studies have already examined the hematopoietic recovery after irradiation but paid with very little attention to the bone marrow microenvironment. Nonetheless previous studies in a murine model of reversible radio-induced bone marrow aplasia have shown a significant increase in alkaline phosphatase activity (ALP) prior to hematopoietic regeneration. This increase in ALP activity was not due to cell proliferation but could be attributed to modifications of the properties of mesenchymal stem cells (MSC). We thus undertook a study to assess the kinetics of the evolution of MSC correlated to their hematopoietic supportive capacities in mice treated with sub lethal total body irradiation. In our study, colony-forming units-fibroblasts (CFU-Fs) assay showed a significant MSC rate increase in irradiated bone marrows. CFU-Fs colonies still possessed differentiation capacities of MSC but colonies from mice sacrificed 3 days after irradiation displayed high rates of ALP activity and a transient increase in osteoblastic markers expression while pparγ and neuropilin-1 decreased. Hematopoietic supportive capacities of CFU-Fs were also modified: as compared to controls, irradiated CFU-Fs significantly increased the proliferation rate of hematopoietic precursors and accelerated the differentiation toward the granulocytic lineage. Our data provide the first evidence of the key role exerted by the balance between osteoblasts and adipocytes in spontaneous bone marrow regeneration. First, (pre)osteoblast differentiation from MSC stimulated hematopoietic precursor's proliferation and granulopoietic regeneration. Then, in a second time (pre)osteoblasts progressively disappeared in favour of adipocytic cells which down regulated the proliferation and granulocytic differentiation and then contributed to a return to pre-irradiation conditions.

Figure 1. Percentages and absolute numbers of colony forming units-fibroblasts (CFU-Fs) in sublethally (4Gy) irradiated bone marrows.

A: 10 days-cultured CFU-Fs colonies stained with May Grünmwald Giemsa (a: original magnification ×50 and b: original magnification ×200). B: Proportion of CFU-Fs in irradiated bone marrows (black columns) compared to controls (grey columns).C: Absolute number of CFU-Fs in irradiated bone marrows (black columns) compared to controls (grey columns). Data are presented as mean values ± SEM of at least three independent experiments. ***, p<.001 ; **, p<.01 ; *, p<.05 as assessed by one way analysis of variance (ANOVA).

Figure 2. Alkaline phosphatases (ALPs) in CFU-Fs.

A: Morphologic observation of ALP activity (in blue) both in in situ bone marrows from controls (Aa) or irradiated (Ab) mice (day 3 after irradiation) and in 10 days-cultured CFU-Fs from controls (Ac) or irradiated (Ad) mice (original magnification ×200). B: Comparison of the ALP activity quantified by colorimetric reaction in 10 days-cultured CFU-Fs from controls and day-3 irradiated mice. C: Expression of Tissue non specific alkaline phosphatase (TNSALP) and Plasma membrane calcium ATPase (PMCA) in 10 days-cultured CFU-Fs : CFU-Fs are positive for TNSALP (Ca) and PMCA (Cb) isoforms as tested by immunohistochemistry (original magnification: ×200). mRNAs for the two isoforms are increased soon after irradiation (Cc : TNSALP mRNA and Cd : PMCA mRNA). Data are presented as mean values ± SEM of at least three independent experiments. ***, p<.001; **, p<.01 ; *, p<.05 as assessed by one way analysis of variance (ANOVA).

Figure 3. Expression of osteogenic and adipocytic markers in CFU-Fs after irradiation.

A: Kinetics of expression of osteoblastic markers mRNAs (Aa: Runx2, Ab: Osteopontin and Ac: Osteocalcin) after irradiation. B: Kinetics of expression of the adipocytic marker PPARγ mRNA after irradiation. C: Kinetics of expression of the adipocytic protein PPARγafter irradiation. D: Kinetics of expression of the osteoblastic protein Runx2after irradiation. Data are presented as mean values ± SEM of at least three independent experiments. ***, p<.001; **, p<.01; *, p<.05 as assessed by one way analysis of variance (ANOVA).

Figure 4. Evolution of white blood cells (WBCs) numbers, of neutrophils, lymphocytes and platelets in peripheral blood after a 4Gy total body irradiation.

A: Evolution of WBCs in peripheral blood after irradiation. B: Number of circulating cells (WBC, neutrophils, lymphocytes and platelets) at different time points after irradiation. Data displayed as mean values ± SEM of at least three independent determinations. ***, p<.001; **, p<.01; *, p<.05 as assessed by one way analysis of variance (ANOVA).

Figure 5. Evolution of the percentage of hematopoietic precursors in sublethally (4Gy) irradiated bone marrows.

A: Granulocytes (Gr-1+), B: Monocytes (CD11b+), C: Lymphocytes (CD45R/B220+). Data are presented as mean values ± SEM of at least three independent experiments. ***, p<.001 ; **, p<.01 ; *, p<.05 as assessed by one way analysis of variance (ANOVA).

Figure 6. Proliferation and survival of hematopoietic precursors in co-culture with CFU-Fs from irradiated (grey) and non-irradiated (black) bone marrows.

A: Proliferation of hematopoietic precursors tested by ³H incorporation. B: Survival of hematopoietic precursors tested by Annexin-PI labeling. Data are presented as mean values ± SEM of at least three independent experiments. **, p<.01 as assessed by one way analysis of variance (ANOVA).

Figure 7. Differentiation of hematopoietic precursors in direct contact with CFU-Fs from irradiated (black) and non-irradiated (grey) bone marrows.

A: Immunophenotyping shows a myelo-monocytic (CD11b), a granulocytic (Ly-6G) and a B-lymphoid (CD45R/B220) engagement of hematopoietic precursors in lineages. B: The differentiation of mature neutrophils during the co-culture assay occurs earlier when precursors are co-cultured on “day 3” CFU-Fs. C: Inhibition of the contact in transwell conditions induces a significant reduction in the differentiation of the Gr1+ (Ca) and CD11b+ (Cb) lineages. Data are presented as mean values ± SEM of at least three independent experiments. **, p<.01; *, p<.05 as assessed by one way analysis of variance (ANOVA).

Figure 8. Expression of neuropilin-1 in CFU-Fs.

A: Immunohistochemistry for Neuropilin-1 expression in irradiated mice CFU-Fs (magnification ×200). B: Evolution of neuropilin-1 mRNAs in in vitro-irradiated CFU-Fs. Data are presented as mean values ± SEM of at least there independent experiments. **, p<.01; *, p<.05 as assessed by one way analysis of variance (ANOVA).

Figure 9. SDF-1 expression in irradiated bone marrows and CFU-Fs.

A: Labelling with anti-sdf1 antibodies is more intense in bone marrows from day-1 irradiated mice (Ab) compared to control ones (Aa) (original magnification ×200). B: Elisa quantification of sdf-1 in the supernatant of irradiated CFU-Fs showed an increase in sdf-1 until 4 hours after irradiation. C: The level of mRNA expression is not significantly different in control and irradiated CFU-Fs. Data are presented as mean values ± SEM of at least three independent experiments. ***, p<.001 as assessed by one way analysis of variance (ANOVA).