Expression of the G-CSF Receptor in Monocytes Is Sufficient to Mediate Osteoblast Suppression and HSPC Mobilization by G-CSF in Mice.

Abstract 563

Granulocyte-colony stimulating factor (G-CSF) is the most common agent used to mobilize hematopoietic stem/progenitor cells (HSPCs) for stem cell transplantation. We previously showed that HSPC mobilization by G-CSF is associated with significant changes in the bone marrow microenvironment. Most notably, there is a marked loss of mature osteoblasts in the bone marrow following G-CSF treatment. These findings raise the possibility that the hematopoietic compartment plays a hitherto unrecognized role in maintaining bone homeostasis. Bone marrow transplantation experiments demonstrated that G-CSF does not directly suppress osteoblast function but acts through a transplantable hematopoietic intermediary. The identify of this hematopoietic intermediary and the signals generated by these cells that suppress osteoblasts are unknown. Earlier work from our lab suggested that mature neutrophils and B- and T-lymphocytes are dispensable for G-CSF-induced osteoblast loss. We therefore hypothesized that bone marrow monocytes play a role in regulating osteoblasts both during G-CSF treatment and at steady state. Consistent with this hypothesis, we previously reported that purified murine monocytes/macrophages stimulated the growth and terminal differentiation of osteoblasts in vitro. To further test this hypothesis, we generated a transgenic mouse in which G-CSFR expression is largely confined to monocytic cells. Specifically, we expressed the G-CSFR and GFP under control of the CD68 promoter. CD68 is a glycoprotein whose expression is restricted to the surface of cells of the monocytic lineage. Two founder lines were identified and backcrossed to G-CSFR^-/- mice, generating mice (termed CD68:G-CSFR) in which the G-CSFR expression is predicted to be restricted to monocyte lineage cells. Consistent with this prediction, we detected G-CSFR expression on monocytes but not granulocytes or lymphocytes from CD68:G-CSFR mice. At baseline, CD68:G-CSFR mice are neutropenic but had normal levels of circulating monocytes, B and T cells, and colony-forming cells (CFU-C). Wild-type, G-CSFR^-/-, and CD68:G-CSFR mice were treated with G-CSF for 7 days and HSPC mobilization and osteoblast activity in the bone marrow were assessed. As expected, G-CSFR^-/- mice had no discernible response to G-CSFR. In contrast, the CD68:G-CSFR mice had a strong mobilization response to G-CSF, with an increase in circulating CFU-C (CFU-C per ml ± SEM: 6,400 ± 1,008) that was comparable to wild-type mice (5,500 ± 2,000). Osteoblast activity was assessed by quantitative RT-PCR for osteocalcin and CXCL12 on bone marrow aspirates. Consistent with previous reports, in wild-type mice, G-CSF treatment resulted in a 38-fold decrease in osteocalcin expression [relative osteocalcin mRNA to beta-actin ± SEM: 18.1 ± 9.08 (baseline) and 0.47 ± 0.21 (G-CSF)] and a 4.9-fold decrease in CXCL12 expression [5.84 ± 3.12 (baseline); 1.19 ± 0.45 (G-CSF)]. A similar 81-fold decrease in osteocalcin (0.22 ± 0.09) and 7.5-fold decrease in CXCL12 (0.77 ± 0.18) expression were observed in CD68:G-CSFR mice after G-CSF treatment. Together, these data demonstrate that G-CSFR expression in monocytes is sufficient to generate the signals required for the suppression of osteoblast activity and HSPC mobilization in mice.