The cytotoxicity of arsenic trioxide to normal hematopoietic progenitors and leukemic cells is dependent on their cell-cycle status

Arsenic trioxide (ATO) is a novel agent to treat acute promyelocytic leukemia (APL). ATO can degrade chimeric PML-RAR proteins and induce apoptosis in various cancer cells. However, its effects on primary hematopoietic CD34(+) have not been examined. In this study, we compared the effects of ATO on HL60 leukemic cells and primary umbilical cord blood (UCB) CD34(+) cells. HL60 cells and UCB CD34(+) cells were cultured with different concentrations of ATO for up to three weeks and examined for changes of cell cycle. We found that ATO (less than or equal to5 muM) caused prolongation of G(1)/S and G(2)/M phase in a dose-dependent manner. The percentage of cells in G(2)/M increased significantly (from 8.6 to 53.8%). High-dose ATO (greater than or equal to25 muM) caused non-specific cell death in HL60 cells without any changes in cell cycle. In contrast to HL60 cells, UCB CD34(+) cells were more resistant to high-dose ATO and most ATO-resistant CD34(+) cells remained in G(0)/G(1) phase. Primary cells that were resistant to ATO were rich in CD34(+) cells. We further show that the ATO resistance was not related to the expression of P-glycoprotein (MDR-1). Our results suggest that the resistance to ATO in primitive UCB CD34(+) cells is most likely related to its cell-cycle status. These results could be useful to design treatments for non-APL malignancies and to enrich hematopoietic stem cells in clinically applicable settings.