Generation of a BCR-ABL-Induced Leukemia Model in Osteopontin-Deficient Mice.

Osteopontin (OPN) is a secreted phoshoprotein playing multiple roles in cell migration, apoptosis and angiogenesis with evidence of increased expression in several types of epithelial tumors. Recently, OPN has also been shown to be a key regulator of the migration of hematopoietic stem cells towards the osteoblastic niche where it participates to the maintenance of the quiescence of stem cells after binding to integrins. We have previously shown that OPN is upregulated by BCR-ABL in a tyrosine-kinase-dependent manner in a TET-regulated BCR-ABL expressing cell line. To further study the potential role of OPN in BCR-ABL-associated leukemogenesis, we have used a retrovirus mediated BCR-ABL gene transfer strategy in OPN-deficient mice (OPN−/−). Bone marrow stem cells were harvested from OPN−/− and wild type OPN+/+ (C57Bl/6) mice after 5-FU treatment at day-5 and retrovirally transduced by a MIGR-p210 BCR-ABL vector in the presence of cytokines for 3 days. Retrovirally transduced marrow cells were transplanted into lethally irradiated OPN−/− mice in the presence of appropriate irradiation controls. Transplanted animals were followed regularly by blood counts, appearance of clinical disease and the results were compared to those obtained in OPN+/+ mice transplanted in the same conditions. In preliminary experiments, we have established that BCR-ABL-transduced OPN+/+ marrow can generate leukaemia in the background of OPN−/− mice with similar efficiencies (n=3) as in the background of OPN+/+ mice (n=3). In following experiments we have observed excessive mortality after irradiation (9–9.5 Gys) and established the optimal protocol pour induction of leukaemia in OPN−/− mice transplanted with OPN−/− cells. We have found that co-transplantation of 5.104 retrovirally transduced BM cells in the presence of 1.5 105 untransduced bone marrow allows reproducibly leukemia induction. The majority of OPN−/− mice (20/24, n=2 experiments) transplanted by the use of this protocol developed hyperleucocytosis between day 21–35 after transplantation. The latency of leukaemia did not differ between wild type and OPN−/− animals. At the time of sacrifice, all mice were found to have splenomegaly (500–950 mg). Both marrow and spleen cells had massive infiltration with GFP+ cells (40–80 %, marrow; 40–90%, spleen). At phenotypic analysis, OPN−/− leukemic cells were found to express at variable degrees Gr1, CD11b, Ter119, CD19, suggesting the establishment of a multilineage disease in the background of OPN−/− marrow. The phenotypic analysis leukemic cells from OPN+/+ mice was found to be similar. Culture of spleen cells from of a leukemic OPN−/− mouse gave rise to a GFP+ cell line expressing c-kit (97%) with negativity of Gr1, B220, Ter119 expression, suggesting its mast cell phenotype. Interestingly, this cell line has exquisite sensitivity to imatinib mesylate, the cytotoxicity of which is reduced by IL-3 but not by stem cell factor. In summary, these results demonstrate that OPN, a transcriptional target of BCR-ABL oncogene, is not required for BCR-ABL-induced murine leukemogenesis. It remains to be determined if it plays a role in the stem cell quiescence of BCR-ABL-expressing primitive cells in the osteoblastic niche and their resistance to BCR-ABL-targeted therapies. The OPN−/− murine CML modelthat we have established will be used to explore these questions as well as the homing potential of leukemic cells in the presence or in the absence of OPN.