Circulating osteogenic cells: characterization and relationship to rates of bone loss in postmenopausal women

There is increasing evidence that osteogenic cells are present not only in bone marrow (BM) but also in peripheral blood (PB). Since staining for alkaline phosphatase (AP) identifies osteoprogenitor cells in BM, we sought to further characterize BM versus PB hematopoietic lineage negative (lin-)/AP+ cells and to compare gene expression in PB lin-/AP+ cells from postmenopausal women undergoing rapid versus slow bone loss. PB lin-/AP+ cells were smaller than their BM counterparts, and both were negative for the pan-hematopoietic marker, CD45. BM and PB lin-/AP+ cells were capable of mineralization in vitro. Using whole genome linear amplification followed by quantitative polymerase chain reaction (QPCR) analysis, we found that relative to the BM cells, PB lin-/AP+ cells expressed similar levels of a number of key osteoblast marker genes (runx2, osterix, osteopontin, OPG, periostin), consistent with the PB cells being in the osteoblastic lineage. Importantly, however, compared to the BM cells, PB lin-/AP+ cells expressed lower levels of mRNAs for AP, type I collagen, and for a panel of proliferation markers, but higher levels of osteocalcin, osteonectin, and PTHR1 mRNAs, as well as those for RANKL and ICAM-1, both of which are important in supporting osteoclastogenesis. Using microarray followed by QPCR analysis, we further demonstrated that, compared to postmenopausal women undergoing slow bone loss, PB lin-/AP+ cells from women undergoing rapid bone loss expressed lower levels of mRNAs for hydroxyprostaglandin dehydrogenase, interferon regulator factor 3, Wnt1-induced secreted protein 1, and TGFbeta2, but higher levels of the Smad3 interacting protein, zinc finger DHHC-type containing 4 and col1alpha2. These data thus demonstrate that while PB lin-/AP+ cells express a number of osteoblastic genes and are capable of mineralization, they are a relatively quiescent cell population, both in terms of cell proliferation and matrix synthesis. However, their higher expression of RANKL and ICAM-1 mRNAs as compared to BM lin-/AP+ cells suggests a role for the PB lin-/AP+ cells in regulating osteoclastogenesis that warrants further investigation. Our study also provides "proof-of-concept" for the use of PB lin-/AP+ cells in clinical-investigative studies, and identifies several pathways that could potentially regulate rates of bone loss in postmenopausal women.