Arsenic trioxide (ATO) is an inorganic arsenic derivative that is effective in patients with relapsed acute promyelocytic leukemia. It is being investigated as therapy for other cancers but the risk/benefit ratio for its use still has to be determined due to its significant side effects. In contrast, organic arsenic derivatives are known to be much less toxic. We have synthesized a series of eight organic arsenic lipid derivatives and have tested them in NIH 60 cell line screen. Based on its activity, we selected GMZ27 [3-S-di(n-propyl)arsino-3-thio-1,2-propanediol] for further study in our laboratory and have confirmed its potent activity against human acute myeloid leukemia cell lines HL60 and NB4, which was significantly higher than that of arsenic trioxide. The IC50 (concentration that kills 50% of the cells) in MTS assay is 0.5 and 0.9 μm for GMZ27. Assessment of the mechanisms of action of GMZ27 in cell lines has shown that the GMZ27 was more potent inducer of superoxide than ATO. GMZ27 caused dissipation of mitochondrial transmembrane potential, cleavage of caspase 9, caspase 3 activation, PARP cleavage, and compromise in cell membrane integrity. However, its treatment also resulted in caspase 8 cleavage, suggesting that it affects both intrinsic and extrinsic apoptotic pathway. Its activity was related to the level of glutathione in the leukemic cells as pretreatment of the cells with BSO, which depletes intracellular glutathione, results in cells sensitivity to GMZ27. On the other hand, pretreatment with DDT, which increases intracellular glutathione, results in their resistance to GMZ27. GMZ27 had no effect on cells maturation and differentiation, and cell cycle. GMZ27 when tested against healthy donor mononuclear cells in a colony forming assay showed significantly less toxicity than arsenic trioxide. In vivo toxicity testing in Swiss Webster mice showed LD50 (dose that kills 50% of mice) to be 100 mg/kg for GMZ27, comparing to 10 mg/kg for ATO. In conclusion, organic arsenic lipid derivatives, and in particular GMZ27, may have more potent antileukemic activity and significantly less toxicity in vivo and in vitro than ATO and therefore further development of these medications is warranted.

Author notes

Corresponding author

Sign in via your Institution