Cytotoxicity of ascorbate and other reducing agents towards cultured fibroblasts as a result of hydrogen peroxide formation

Several types of cultured fibroblasts, including chick embryo, human and mouse, were killed by the addition of sodium ascorbate at final concentrations of 0.05-0.25 mM to cultures at the time of inoculation or to attached cells. Ascorbate did not affect the attachment of cells to the substratum. The effect on chick embryo fibroblasts was visible by four hours and by six hours almost all cells had swelled and were becoming detached. By 24 hours detached cells had either lysed or become crenated in appearance. Other end-diol reducing agents and also glutathione and cysteine were effective while gulonolactone, a non-reducing analogue of ascorbate, was ineffective. Preincubation of medium containing ascorbate but no cells, conditions which result in degradation of the vitamin, led to loss of toxicity, indicating that a degradation product was not the lethal agent and that a component of the medium was not converted to a lethal substance. The lethal effect of both ascorbate and glutathione was prevented by the addition of catalase to the medium suggesting that H2O2 formed by intracellular reactions and then excreted into the medium was the cytotoxic agent. This conclusion was supported by the findings that 0.05 mM H2O2 added to chick embryo fibroblasts was lethal and that the effect of this compound on cellular morphology was almost identical to that of ascorbate.