Drug resistance of stem-like cells of BRCA1/p53 associated breast cancer in conditional knockout mice

5722 Cancer stem cells have been postulated to be involved in cancer recurrence and chemoresistance. We have observed that mammary gland tumors, in a mouse model that carries a tissue specific disruption of p53 and BRCA1, recurred after complete regression following cisplatin (CDDP) treatment. We sought to examine the possible involvement of CDDP-resistant cancer stem cells in this phenomenon, and to characterize these cells. We isolated potential cancer stem cells using FACS sorting based on CD24 and CD44 cell surface marker expression. We assayed for their ability to self-renew in culture and to initiate tumor growth in vivo. Our investigation revealed that these cells were able to grow in serum-free media as loose clusters of suspension cells. They were not able to form mammospheres, an observation that has recently been reported with human samples. Colony forming assays performed on irradiated NIH3T3 feeder layers showed that the CD24-positive cells resulted in ten-fold more colonies compared to CD44-positive cells. This observation suggested that CD24-positive cells were more likely to be the stem cells. Transplantation of the CD24-positive cells into cleared fat pads of immunodeficient mice resulted in tumor regeneration. CDDP treatment of the tumors resulted in partial regression, suggesting an increase in CDDP resistant cells in the regenerated tumor. This response appeared to be dependent on the number of cells injected, since a five-fold increase in cell number resulted in almost complete resistance to CDDP. FACS analysis of the regenerated tumors revealed a phenotypically diverse population of cells expressing a wide range of CD24 levels. These data are suggestive of the CD24-positive cells’ stemness, which reflects their ability to generate cells of diverse phenotypes. A large proportion of the CD24-positive cells was found to express high levels of CD29, another putative stem cell marker. A secondary transplantation, using 5000 of these CD24-positive cells, also resulted in tumor regeneration. Tumor growth increased tremendously in a short time period following the secondary cell injection. CDDP treatment had no effect on these tumors. FACS analysis of cells harvested from the tumors also showed regeneration of cells with a wide range of CD-24 expression levels. Most of the CD24-positive cells in this population were also CD29-positive. Our study demonstrates that CDDP resistance observed in this mouse model may involve drug-resistance cancer stem cells expressing high levels of CD24 and CD29. These data may provide useful information towards improved therapies targeting a small subpopulation of drug-resistant cancer stem cells.