Paclitaxel is able to cause cell death through the induction of apoptosis. Cell death characteristics for docetaxel have not yet been described in detail. We investigated four unselected human ovarian cancer cell lines for the sensitivity to a 1hr exposure to docetaxel and calculated the concentrations inhibiting 50% (IC(50)) and 90% (IC(90)) of cell growth. Of the cell lines A2780, H134, IGROV-1 (all wild-type p53) and OVCAR-3 (mutant, mt p53) A2780 was most sensitive and OVCAR-3 least sensitive. Equitoxic drug concentrations representing IC(90) values (25-510nM) were applied for 1hr to measure cell cycle distribution, DNA degradation, and to count apoptotic cell bodies and cells with multifragmented nuclei at various time-points after drug exposure. H134, IGROV-1 and OVCAR-3 showed a continued mitotic block up to at least 72hr and prolonged presence of cells with multifragmented nuclei. High percentages of apoptosis were calculated at 48hr and at later time-points. In contrast, A2780 cells accumulated in the S-phase of the cell cycle and apoptosis was hardly present. The changes in the expression levels of p53, p21/WAF1, Bax and Bcl-2, were not predictive for docetaxel-induced apoptosis. Caspase-3 activation occurred only in cells with accumulation in the G2/M phase starting as early as 8hr in OVCAR-3. Prolonged Bcl-2 phosphorylation was evident in OVCAR-3, visible at 24hr in H134 and IGROV-1, while this phenomenon did not occur in A2780. The mitogen-activated protein kinase pathway (JNKs/SAPKs or c-Jun N-terminal kinases/stress-activated protein kinases, JNK1/2; extracellular response kinase, ERK1/2; p38) did not seem to be directly involved in Bcl-2 phosphorylation or apoptosis. We conclude that docetaxel is able to activate caspase-3, induce Bcl-2 phosphorylation and apoptosis in cells that show a prolonged G2/M arrest, but cells may also die by a caspase-3-independent cell death mechanism.