A Novel in-Vivo Model to Examine Homing of Multiple Myeloma Cells in Postnatal, Inducible, and Reversible Loss of Mature Osteoblasts

The initiation of Multiple myeloma (MM) coincides with a decrease in osteoblasts and an increase in osteoclasts and adipocytes. However, the contribution of osteoblasts to the initiation and progression of MM remains mostly unknown. In-vitro studies have shown that co-cultures with MC3T3 murine osteoblast cell-line induced quiescence in 5TGM1 murine myeloma cells. In SCID or C57BL/KaLwRijHsd mice, the fluorescently labeled MM cell lines, injected through tail-vein, co-localized with endosteal osteoblasts and were found to be quiescent. We previously demonstrated that increased osteoblast numbers using Activin A inhibition in SCID mice with human-origin MM.1S cells resulted in inhibition of MM growth (Vallet et al., PNAS 2010). Here we aimed to develop a genetic in-vivo model for a better understanding of the interactions between osteoblasts and MM cells.

Osteolineage cells develop through the stages of pre-osteoblasts, committed osteoblasts, mature osteoblasts, and osteocytes; each stage governed by a distinct molecular signature and function. To delineate these differences, individual osteoblast populations were FACS-sorted from long bones of Osterix-GFP+ (Osx+) mice (pre-osteoblasts), Collagen 2.3-GFP+ (Col2.3+) mice (committed osteoblasts), and Osteocalcin-YFP+ (OCN+) mice (mature osteoblasts). Co-cultured of these osteoblast populations with 5TGM1 MM cells showed that the OCN+ mature osteoblasts suppressed MM proliferation the most. Therefore, we focused our studies on the role of mature osteoblasts in MM homing and engraftment.

We next generated mice in which mature osteoblasts were postnatally deleted in an inducible and reversible manner to better understand the molecular mechanisms of engraftment, proliferation, and migration of 5TGM1 injected intra-tibialy. To achieve this, mice carrying floxed diphtheria toxin receptor (DTR) alleles were mated with mice expressing Cre-recombinase driven by the osteocalcin promoter (OC-Cre). This led to the expression of DTR in mature osteoblasts (OC-Cre/iDTR) only. The control mice were littermates lacking the OC-Cre allele (iDTR). The OC-Cre/iDTR mice were indistinguishable from the controls until treated with diphtheria toxin (DT). We chose to induce mature osteoblast-deficiency at 8-weeks to allow skeletal maturation which is assumed to be completed by 6-7weeks of age. To induce postnatal deletion of mature osteoblasts, the OC-Cre/iDTR and control mice both were treated with 50 µg/Kg DT once a week, beginning at 8-weeks of age. Micro-CT analysis showed a significant increase in cortical porosity within 1-week after DT injection. 8-weeks of DT treatment significantly reduced trabecular bone fraction (BV/TV), trabecular numbers (Tb.N), and bone mineral density (BMD) with a significant increase in trabecular spacing (Tb.Sp). Immunohistochemistry for osteocalcin showed rapid loss of mature and endosteal osteoblasts (N.Ob/T.Ar). This was accompanied with a marked decreased in serum sclerostin and serum osteocalcin levels suggesting reduced osteocytes and mature osteoblasts, respectively. Importantly, serum CTX levels or numbers of osteoclasts (N.Oc/T.Ar) were unchanged. To study MM engraftment and progression, 3x10⁶ 5TGM1 luciferase eGFP positive (5TGM1-Luc-GFP) MM cells were injected into the tibia of OC-Cre/iDTR and control iDTR mice followed by weekly injection of DT for 8-weeks. Flow cytometry analysis showed a 4-fold increase in the 5TGM1-GFP cells in the BM from OC-Cre/iDTR mice compared to controls. Bioluminescence imaging (BLI) for 5TGM1-Luc-GFP cells at 4-weeks showed 600-fold increase in the OC-Cre/iDTR mice compared to controls. Interestingly, by 8-weeks, the BLI imaging showed 5TGM1-Luc-GFP cells in other long-bones of OC-Cre/iDTR mice but not the controls.

These data show that MM cells engraft and proliferate rapidly in the absence of mature osteoblasts in-vivo. Moreover, in the absence of mature osteoblasts, MM cells have a propensity to migrate to other long bones. These data further suggest that expanding the mature osteoblast niche may provide novel therapeutic avenues and reduce disease burden and create an environment for long term tumor control. Importantly, this model will allow us to follow and analyze the sequential engraftment, dormancy, reactivation, proliferation, and migration of myeloma cells, and evaluate the effects of osteoanabolic and anti-myeloma therapies.