Anthracyclines are among the most powerful drugs used for the treatment of leukemia, however their use has been associated with cardiotoxicity. Reactive oxygen species (ROS) are generated in both cancer and normal cells after anthracycline exposure and have been implicated in both early and late onset cardiotoxicity. Counteracting this ROS generation are intracellular antioxidants such as the ubiquitous antioxidant glutathione (GSH), levels of which are depleted upon anthracycline exposure. Basal expression of GSH pathway components and other antioxidants vary greatly between different cell types. Due to this differential expression of cellular antioxidants in cardiomyocytes versus leukemia cells, we posit that anthracyclines exert distinct effects on oxidative stress and consequent apoptosis induction in leukemia cells and nontransformed hematopoietic cells (PBMC) relative to cardiomyocytes. As a result, we expect potentially varied mechanisms of cell death induction in these cell lines after anthracycline treatment. To test this hypothesis, the acute leukemia cell lines Jurkat and ML-1 and the cardiomyocyte line H9C2 were used. Dose responses with the anthracyclines, doxorubicin and daunorubicin, were carried out and trypan blue exclusion and propidium iodide staining followed by flow cytometry were used to assess viability and DNA fragmentation respectively. Cardiomyocytes had a 25-150 fold higher IC50 value than the acute leukemia cell lines, indicating selectivity. To assess whether apoptosis was induced by anthracyclines, caspase 3 activity was measured and found to be increased at 24 hours in Jurkat cells which preceded decreases in viability, supporting an apoptotic mechanism of cell death. GSH levels also decreased markedly after 24 hours of treatment with anthracyclines in this cell line, however, a pan-caspase inhibitor did not block GSH depletion, indicating that these events occur independent of each other. To evaluate whether antioxidants conferred protection against loss of viability in all cell types, cells were pretreated for at least 30 minutes with antioxidants and then treated with doxorubicin and daunorubicin for 24 hours. Antioxidants used were N-acetylcysteine (NAC, a GSH precursor and amino acid source), GSH ethyl ester (cell permeable form of GSH), tiron (free radical scavenger) and trolox (a water soluble form of vitamin E). GSH ethylester did not prevent cytotoxicity of anthracyclines in acute leukemia lines or cardiomyocytes. Therefore boosting GSH levels in leukemia cells does not reverse cytotoxicity. Trolox, however, did block anthracycline induced cell death in ML-1 cells, suggesting that vitamin E supplementation would counteract leukemia cell specific effects of anthracyclines on AML cells. Tiron protected PBMC from doxorubicin cytotoxicity but did not protect leukemia cells or cardiomyocytes, hinting at a protective strategy for normal non-leukemia blood cells. Interestingly, NAC did not interfere with the cytotoxic effects of anthracyclines on acute leukemia cells or PBMC, but protected H9C2 cells from daunorubicin cytotoxicity. Taken together, these data reveal differential protective effects of antioxidants in cardiomyocytes and PBMCs relative to ALL and AML cells. Our work indicates that NAC can protect cardiomyocytes without interfering with anthracycline cytotoxicity in acute leukemia cells. In humans, one randomized control trial tested the addition of NAC to doxorubicin therapy, detecting no evidence of cardioprotective activity by chronic administration of NAC. However, the schedule used for administration of NAC in that study may not have been optimal, and biomarkers for oxidative stress reduction by NAC were not incorporated into the trial. Previously, other antioxidants have been used with very limited clinical success and possible contributing factors include inadequate sample size, choice of agent, dose used, duration of intervention and the lack of biomarker endpoints. Designing a cardioprotective and antioxidant strategy with attention to these factors may prove to be efficacious in protecting cardiac cells without interfering with the antitumoral effect of anthracyclines. To this end, our data suggests that trolox and vitamin E analogues should not be used in acute leukemia as they may interfere with the cytotoxic action of anthracyclines but NAC or cysteine may be used as cardioprotectants.


No relevant conflicts of interest to declare.

Author notes


Asterisk with author names denotes non-ASH members.