We have previously reported that zoledoronic acid (ZOL) augmented the in vivo effect of imatinib in a murine chronic myeloid leukemia (CML) model (Blood 2003). ZOL alone induces apoptosis in leukemic cells in vitro by inhibiting prenylation of the Ras-related proteins. In addition to this direct anti-leukemic effect, we hypothesized that ZOL also has some influence in leukemic cells in vivo indirectly by destroying osteoclasts (OCs), which is the primary therapeutic activity of ZOL in osteoporosis patients. Supporting this notion is that by mediating bone resorption, OCs release a variety of cytokines such as IGF- 1, TGF-β, etc. that have accumulated in the bone matrix. It has been reported that OCs play an important role in bone metastasis of solid tumor, especially in cancer stem cells. However, little is known about the role of OCs in leukemia. Therefore, we investigated it in vitro and in vivo. For this purpose, we established an in vitro osteoblasts (OBs) and OCs co-culture system. The stable co-culture system that we developed includes collagen gel and murine primary OBs and OCs. In addition, murine femoral bone sections were sometimes added to this culture system so that the OCs could release the cytokines from the bone matrix. Thus, the collagen gel and OBs were placed in 12-well plates with and without bone sections and/or OCs. The transwell chambers over the wells then received 1×104 Ba/F3 cells that had been transfected with wild type bcr-abl (Ba/F3/bcr-abl cells). OBs markedly enhanced the growth of Ba/F3/bcr-abl cells in this indirect contact coculture system whereas the presence of both OBs and OCs slightly suppressed cell growth. Intriguingly, when bone sections were added (OBs+OCs+bone), Ba/F3/bcr-abl cell proliferation was significantly suppressed compared to the effect of OBs alone or OBs+OCs (Figure). Cell cycle analysis revealed that the G0/G1 population was increased in Ba/F3/bcr-abl cells co-cultured with OBs+OCs+bones. We also observed that the p27 protein levels of Ba/F3/bcr-abl cells increased upon co-culture with OCs or OCs+bones, similar to their response to treatment with purified TGF-β. We performed ELISAs to determine the concentrations of cytokines in the supernatants of co-cultured OBs and OCs. There were higher levels of TGF-β1 in the OBs+OCs+bones supernatant than in the OBs+OCs supernatant. Furthermore, OBs produced high levels of IGF-1. These findings suggest that OBs and OCs affect the proliferation and the cell cycle arrest of leukemic cells by releasing soluble factors, respectively. To more comprehensively elucidate the roles OCs play in leukemia cells in vivo, we used reveromycin A (RM-A) which inhibits bone resorption by specifically inducing apoptosis in OCs (Woo et al, PNAS 2006). RM-A did not have any in vitro effects on the proliferation of Ba/F3/bcr-abl cells. Thus, we could know the unalloyed role of OCs in leukemia with RM-A compared with ZOL which inhibited directly both OCs and leukemic cells. Our preliminary data show that RM-A suppresses the engraftment of inoculated Ba/F3/bcr-abl cells to nude mice. We also present data from ongoing studies showing the effect of RM-A on leukemic cells in murine models. These findings suggested that OCs may be an important constituent of leukemia stem cell niche and destruction of OCs by either ZOL or RM-A is a novel strategy for leukemia treatment.

Disclosures: No relevant conflicts of interest to declare.

Author notes

Corresponding author