Depletion of cellular glutathione by N,N'-bis(trans-4-hydroxycyclohexyl)-N'-nitrosourea as a determinant of sensitivity of K562 human leukemia cells to 4-hydroperoxycyclophosphamide.

N,N'-Bis(trans-4-hydroxycyclohexyl)-N'-nitrosourea (BHCNU) is a nitrosourea which has carbamoylating but not alkylating activity. It has been shown to carbamoylate and inactivate glutathione reductase thereby reducing the intracellular levels of glutathione (GSH). Since GSH depletion by buthionine-S,R-sulfoximine potentiates the cytotoxicity of cyclophosphamide, with a corresponding increase in DNA cross-linking, we have investigated the potential interaction between BHCNU and cyclophosphamide. Treatment of K562 human leukemia cells with 15 microM BHCNU for 1 h resulted in depletion of glutathione to 40% of control values, without significant reduction of cell viability. Subsequent treatment with 10 microM 4-hydroperoxycyclophosphamide (4-HC), a self-activating derivative of cyclophosphamide, reduced the level of glutathione to less than 20% of control values. BHCNU pretreatment enhanced the cytotoxicity of 4-HC resulting in a dose modification factor of 2.5. Alkaline elution analysis of cellular DNA demonstrated that the level of interstrand cross-linking was 2-fold higher in GSH-depleted cells than in nondepleted cells, and the induction of single strand breaks was markedly increased. These findings demonstrate that BHCNU potentiates the cytotoxicity of 4-HC and suggest that this is due to the increased formation of DNA interstrand cross-links caused by a reduced intracellular conjugation of 4-HC with glutathione which results in an increased binding of 4-HC to DNA targets.

Glutathione depletion as a determinant of sensitivity of human leukemia cells to cyclophosphamide.

The role of glutathione (GSH) as a determinant of cellular sensitivity to the cytotoxic and DNA-damaging effects of cyclophosphamide (CP) was studied in a dual culture system of rat hepatocytes and K562 human chronic myeloid leukemia cells, which have elevated aldehyde dehydrogenase activity with a corresponding insensitivity to activated CP. Exposure of K562 cells to 50 microM DL-buthionine-S,R-sulfoximine for 24 h resulted in a depletion of cellular GSH content to 10% of control values without toxicity. Subsequent 1-h exposure of GSH-depleted cells to activated cyclophosphamide, obtained by incubation of CP with suspension cultures of rat hepatocytes, resulted in a 5-fold potentiation of the cytotoxicity of CP. Alkaline elution analysis of cellular DNA demonstrated that the level of apparent interstrand cross-linking was 3 to 4 times higher in GSH-depleted cells than in nondepleted cells. GSH-depleted cells were, in addition, more sensitive to induction of DNA single strand breaks than nondepleted cells. Depletion of GSH content did not increase cellular sensitivity to the cytotoxicity of phosphoramide mustard. Preincubation of K562 cells with 1 mM cysteine for 4 h resulted in an approximately 60% increase in cellular GSH content, which was accompanied by decreased sensitivity to the cytotoxicity of hepatocyte-activated CP. Exposure of nondepleted cells to clinically relevant concentrations of hepatocyte-activated CP resulted in depletion of cellular GSH content. Replenishment of GSH content in these cells was relatively slow following CP exposure. Acrolein was highly effective at depleting cellular GSH content, whereas phosphoramide mustard had no effect on cellular GSH content. The depletion of GSH by intracellularly released acrolein may be important in the mechanism of cytotoxicity of CP.

Cytotoxicity, DNA cross-linking, and single strand breaks induced by activated cyclophosphamide and acrolein in human leukemia cells.

The in vitro cytotoxicity and mechanism of action of cyclophosphamide (CP) were studied in a dual cell culture system, using rat hepatocytes and K562 human chronic myeloid leukemia cells. Cytotoxicity and DNA damage were measurable in K562 cells using CP concentrations that are clinically attainable. Alkaline elution analysis of cellular DNA demonstrated the presence of concentration- and time-dependent DNA interstrand cross-links, DNA-protein cross-links, and DNA single strand breaks in K562 cells following a 1-h exposure to cyclophosphamide activated by hepatocytes. Hepatocyte-activated CP was 3 to 4 times more potent than phosphoramide mustard with regard to cytotoxicity and induction of DNA interstrand cross-links. Exposure to phosphoramide mustard did not produce single strand breaks, but exposure of K562 cells to acrolein resulted in substantial levels of single strand breaks. The demonstration of acrolein-induced single strand breaks following exposure to activated CP is a novel finding and suggests that acrolein may have a role in the cytotoxicity of CP.