Two controversial issues regarding p53 are whether it is involved in apoptosis induction of tumor cells by a histone deacetylase (HDAC) inhibitor and, given that p53 is indeed involved, which genes of acetylated p53 targets are responsible for giving rise to apoptotic death. We, in the present study, first confirmed that some substantial extent of apoptotic cell death was seen when p53-deficient cells (KATO-III) were transfected with wild-type p53 and treated with sodium butyrate (SB) or trichostatin A. By Western blotting, using specific antibodies, we then demonstrated that residues 320, 373, and 382 lysines of p53 were acetylated in KATO-III cells transfected with wild-type p53 (KATO-III/p53) treated with a HDAC inhibitor. However, as revealed by terminal deoxynucleotidyl transferase-mediated nick end labeling staining, only those KATO-III cells transfected with K320R p53 or K373R p53 became insensitive to the HDAC inhibitor, suggesting that these two residues of p53 may be essential for HDAC inhibitor-induced apoptosis, whereas others such as K382R p53 may not. Furthermore, reverse transcription-PCR demonstrated that among various p53-related proapoptotic genes, expression of PIG3 and NOXA were clearly enhanced by SB treatment in KATO-III/p53 cells but not in KATO-III/K320R or KATO-III/K373R cells. Finally, we revealed that apoptosis could be evoked by SB even in cells where p53 mutations occur at residues other than 320 lysine or 373 lysine (TMK-1 and HSC-39 cells) and that this apoptosis was significantly, although not totally, suppressed by the anti-p53 antisense. It was, therefore, concluded that acetylation of the p53 molecule at residues 320 and 373, giving rise to up-regulation of PIG3 and NOXA, is one of the mechanisms for induction of apoptosis by HDAC inhibitors in cancer cells.