Selenodiglutathione (SDG), the initial metabolite of selenite, is shown to be a more powerful inhibitor of cell growth in vitro than selenite itself. This has been established both with mouse erythroleukaemia (MEL) cells and an ovarian cell line (A2780) which is known to contain wild-type p53. Other seleno-compounds, such as selenomethyl selenocyst-eine (SMS) and dimethyl selenoxide (DMS), which are potent chemopreventive agents and are known to be metabolizedto methylated selenium derivatives directly rather than via SDG, are also growth inhibitory to both MEL and A2780 cells, although less so than SDG or selenite. However, cells growth-inhibited by DMS are more viable than cells growth-inhibitedto the same extent by SDG or selenite, suggesting that the methylated seleno-compounds may inhibit cell growth in a different manner from that of SDG or selenite. Our studies of the mechanism of growth inhibition by SDG, have established two facts. First, SDG induces p53 protein levels in cells that contain wild-type p53 (A2780 cells), suggesting that SDG induces the DNA damage-recognition pathway. Secondly, SDG induces apoptosis in MEL cells, as judged by flow cytometry and formation of nucleosomal DNA ladders. However, since p53 mutations have been found to be targetted events in all MEL cells examined, our evidence suggests that induction of apoptosis by SDG is not absolutely dependent on the p53 response pathway.