MM is the most common cancer metastasizing to bone, with 85–90% of patients developing bone disease. It is unique from other forms of metastatic bone disease, as it is purely lytic. There are a multitude of factors involved in promoting bone disease in MM; however, what remains unanswered is why patients in complete remission from MM are unable to heal their bone lesions. We hypothesize that MM cells exert a permanent change in the bone marrow microenvironment, either through a change in the differentiation potential of mesenchymal stem cells or through stem cell depletion. To investigate this question we developed a new model of MM bone disease in which a herpes-simplex thymidine kinase gene (HSVtk), sensitive to ganciclovir (GCV), is expressed within a MM cell line and injected into the tibia of mice. This allows for eradication of the MM cells without the use of chemotherapy or immunotherapy, minimizes adverse effects on neighboring mesenchymal stem cells and directly involves the bone. The murine MM cell line (5TGM1) was infected, using a lentiviral system, with the tricistronic construct of HSVtk linked to green fluorescence protein (GFP) and blasticidin. Following blasticidin selection, 5TGM1tk cells stably expressing HSVtk and GFP were used for all experiments. GCV dosing curves to determine the dose resulting in 100% eradication were completed. The effects of GCV on both murine hematopoietic cells and mesenchymal stem cells induced to osteoblasts were completed using methylcellulose colony formation and alkaline phosphatase assays. Bystander effects of 5TGM1tk cells on murine hematopoietic cells and murine osteoblasts were completed using murine methylcellulose colony formation, alkaline phosphatase and colony-forming unit fibroblast (CFU-F) assays. In vivo data was obtained following intratibial injection of 5TGM1tk cells (1x105, 5x105, 1x106) or saline into NIH-III mice (3 mice per group, 12 mice total). Mice were evaluated at weekly intervals for tumor and lytic lesion development with plain radiographs and Micro QCT, and osteoblast activity evaluated by alkaline phosphatase analysis. No significant effects of GCV, at doses capable of eradicating the MM cell line (1ug/ml) were observed on murine hematopoietic cells or osteoblasts. No significant effect was observed on hematopoietic colony development when 5TGM1tk cells were cocultured with murine hematopoietic cells. A small bystander effect, approximately 30%, was identified when 5TGM1tk cells were cocultured with murine mesenchymal stem cells induced to osteoblasts in the presence of GCV. Dose-dependent CFU-F suppression was observed with increasing percentages of tumor cells (0–50%), however reversal of suppression was noted when cultured in the presence of GCV. 60% of the mice injected with 5TGM1tk cells developed tumor, and evidence of dose-dependent osteoblast suppression in vivo observed as measured by the alkaline phosphatase assay (p<0.00001). The current model provides us with the capacity to specifically target MM cells without using chemotherapy or immunotherapy. It provides a model to study and dissect the interactions between MM cells and mesenchymal stem cells, within the bone marrow microenvironment, which lead to osteoblast suppression. This model should provide important insights into the mechanisms of osteoblast suppression in MM and help to identify targeted therapies that can reverse this process for patients.

Disclosures: David Roodman is a consultant for Amgen, Novartis, Celgene and Millenium.; Alissa Huston received an Amgen Oncology Institute Hematology/Oncology Fellowship Grant (6/2005-6-2006).; David Roodman is on the Speakers Bureau for Novartis.

Author notes


Corresponding author