Suppression of CXCL12 Production by Bone Marrow Osteoblasts Is a Common and Critical Pathway for Cytokine-Induced Mobilization.

There is strong evidence that CXCL12 (stromal-derived factor-1)/CXCR4 signaling is a key regulator of hematopoietic stem and progenitor cell (HSPC) trafficking in the bone marrow. CXCL12 protein and mRNA expression in the bone marrow are markedly reduced with G-CSF treatment. We and others recently showed that G-CSF treatment results in a marked loss of mature endosteal and trabecular osteoblasts. Since osteoblasts are a major source of CXCL12, this observation provides a potential mechanism by which G-CSF downregulates CXCL12 expression in the bone marrow. To test this hypothesis, we performed RNA in situ studies for CXCL12. These studies show that CXCL12 is highly expressed in endosteal osteoblasts as well as in scattered cells in the bone marrow in untreated mice. In mice treated with G-CSF, there is a near complete loss of CXCL12 signal along the endosteum. Together, these data suggest a model in which G-CSF induced suppression of osteoblasts leads to a decrease in CXCL12 expression in the bone marrow, disrupting CXCR4 signaling and ultimately leading to HSPC mobilization. This model raises several important questions. Many different hematopoietic cytokines can induce HSPC mobilization. Is loss of CXCL12 expression by osteoblasts a common mechanism by which cytokines induce HSPC mobilization? To address this question, we studied HSPC mobilization by Flt3 ligand (Flt3L) and stem cell factor (SCF), two potent mobilizing hematopoietic cytokines. Treatment with Flt3L or SCF resulted in a significant decrease in bone marrow CXCL12 protein and mRNA expression. Moreover, the decrease in CXCL12 expression was accompanied by a loss in trabecular osteoblasts [number osteoblasts per mm bone ± SEM: 5.1 ± 0.1 (control), 3.1 ± 0.4 (Flt3L), 1.8 ± 0.8 (SCF); n=2, p<0.05]. There is considerable evidence implicating other pathways in HSPC mobilization, including upregulation of protease activity in the bone marrow and downregulation of critical adhesion molecules on HSPC and their supporting stromal cells. What is the relative importance of each of these pathways to mobilization in vivo? To address this question, we established chimeras containing CXCR4^−/− hematopoietic cells by transplanting CXCR4^−/− fetal liver cells into irradiated syngeneic recipients. Consistent with previous reports, CXCR4^−/− chimeras display constitutive mobilization of HSPC (number CFU-C/ml blood ± SEM: 12,080 ± 2,904 compared with 120 ± 25 for control chimeras; n=7–9). Surprisingly, G-CSF treatment of these mice did not result in an increase in circulating HSPC (9,511 ± 3,044; n=7–9). This lack of response to G-CSF was not simply due to a loss of mobilizable HSPC in the bone marrow, as treatment with AMD15057, a specific VLA4 antagonist, induced robust mobilization in CXCR4^−/− chimeras (30,600 ± 3,790; n=6–9, p<0.01 compared with G-CSF-treated chimeras). Together, these findings suggest that:

1. suppression of CXCL12 expression by osteoblasts is a common mechanism of cytokine-induced HSPC mobilization;

2. this mechanism represents the principal pathway by which G-CSF mobilizes HSPC, with other pathways (e.g. protease activation) operating upstream or downstream from this pathway; and

3. VLA-4 antagonism, in contrast, provides a distinct mechanism by which HSPC may be mobilized from the bone marrow.