Glutathione (GSH) depletion to approximately equal to 5% of control for 48 h or longer by 0.05 mM L-buthionine sulfoximine (BSO) led to appreciable toxicity for the 66 murine mammary carcinoma cells growing in vitro [L.A. Dethlefsen et al., Int. J. Radiat. Oncol. Biol. Phys. 12, 1157-1160 (1986)]. Such toxicity in normal, proliferating cells in vivo would be undesirable. Thus the toxic effects after acute GSH depletion to approximately equal to 5% of control by BSO plus dimethylfumarate (DMF) were evaluated in these same 66 cells to determine if this anti-proliferative effect could be minimized. Two hours of 0.025 mM DMF reduced GSH to 45% of control, while 6 h of 0.05 mM BSO reduced it to 16%. However, BSO (6 h) plus DMF (2 h) and BSO (24 h) plus DMF (2 h) reduced GSH to 4 and 2%, respectively. The incorporation (15-min pulses) of radioactive precursors into protein and RNA were unaffected by these treatment protocols. In contrast, cell growth was only modestly affected, but the incorporation of [3H]thymidine into DNA was reduced to 64% of control by the BSO (24 h) plus DMF (2 h) protocol even though it was unaffected by the BSO (6 h) plus DMF (2 h) treatment. The cellular plating efficiencies from both protocols were reduced to approximately equal to 75% of control cells. However, the aerobic radiation response, as measured by cell survival, was not modified at doses of either 4.0 or 8.0 Gy. The growth rates of treated cultures, after drug removal, quickly returned to control rates and the resynthesis of GSH in cells from both protocols was also rapid. The GSH levels after either protocol were slightly above control by 12 h after drug removal, dramatically over control (approximately equal to 200%) by 24 h, and back to normal by 48 h. Thus even a relatively short treatment with BSO and DMF resulting in a GSH depletion to 2-5% of control had a marked effect on DNA synthesis and plating efficiency and a modest effect on cellular growth. One cannot rule out a direct effect of the drugs, but presumably the antiproliferative effects are due to a depletion of nuclear GSH with the subsequent inhibition of the GSH/glutaredoxin-mediated conversion of ribonucleotides to deoxyribonucleotides. However, even after extended treatment, upon drug removal, GSH was rapidly resynthesized and cellular DNA synthesis and growth quickly resumed.