Abstract
When cancer cells are resident in bone, they initiate a vicious cycle with osteoclasts (OCs) which perpetuates their growth and aggressive behavior. OCs are critical for the maintenance of the vicious cycle, since they control not only bone destruction associated with cancer, but also the aggressive behavior of tumor cells. It has recently been recognized that tumor cells grow in distant sites because they induce non-malignant cells to establish a “pre-metastatic niche” for tumor cells to later engraft. But nothing is yet known for bone. Primitive bone marrow progenitor cells, called myeloid immune suppressor cells (MISCs), which suppress immune reactivity, are important niche components. MISCs belong to the myelomonocytic lineage with surface markers of Gr-1 and CD11b. We hypothesize that MISCs are precursors of OCs recruited by tumors to assist in the establishment of the vicious cycle. To test this hypothesis, we used the well-characterized 5TGM1 murine myeloma model. 5TGM1-GFP tagged myeloma cells were inoculated via tail vein. The proportion of Gr-1+CD11b+ cells in bone marrow and spleen were assessed by FACS. On week 4 after tumor cell inoculation, %Gr-1+CD11b+ cells were significantly greater in tumor-bearing mice compared with controls (60.9±7.8% vs 37.7±8.6% p<0.05 in marrow; 21.1±4.84% vs 2.4±0.85% p<0.05 in spleen) and paralleled the myeloma burden in bone and spleen. We sorted the Gr-1+CD11b+ cells from the spleens by using magnetic microbeads. MISCs formed multinucleated TRAP positive OCs in medium containing M-CSF (25ng/ml) and RANKL (50ng/ml). The number of OCs derived from tumor MISCs was dramatically greater than those from control mice after 14 day culture (13.4±2.1 vs 1±0.7 per 100x field). MISCs expressed the αv chain of the vitronectin receptor (CD51) and the calcitonin receptor which are specific markers for OCs as they differentiated into multinucleated TRAP+ cells. Only MISCs from tumor-bearing mice and not MISCs from control mice caused resorption pits on dentine discs, demonstrating they were functional OCs. To study the in vivo differentiation of MISCs, we bred lacZ generalized C57 B6 mice with Rag2−/− immune compromised mice to generate lacZ+/− Rag2−/− mice. MISCs were sorted from lacZ+/− Rag2−/− myeloma mice. lacZ positive MISCs were co-injected with 5TGM1 cells to Rag2−/− mice. On day 10 after injection, lacZ+ multinucleated cells could be seen on endosteal surface beneath growth plate by X-gal staining. Those lacZ+ cells were also TRAP+, indicating they were OCs. Treatment of mice with zoledronic acid 100ug/kg s.c. 2/week for 4 weeks reduced % MISCs in tumor bearing mice and impaired the capacity of MISCs to form OCs in vitro (OC# 42.4±4.0 vs 25.6±3.5 per 100x field). Our data suggest that MISCs are increased significantly in marrow and spleen of myeloma-bearing mice and parallel the appearance of lytic bone lesions. These MISCs differentiate avidly and rapidly into functional OCs in vitro as well as in vivo. Zoledronic acid impairs both MISCs, the OC precursors, and mature OCs. These results have a number of implications, including the possibility of reducing bone lesions in myeloma and other malignancies by depleting specific subpopulations of osteoclast precursors.
Disclosures: No relevant conflicts of interest to declare.
Author notes
Corresponding author