We evaluated irinotecan (CPT-11) together with 5-fluorouracil (5-FU) for improved cell growth inhibition with respect to that by either agent alone in the human colon carcinoma cell lines SW620, HT-29 and SNU-C4. Cells were exposed for 24 h to each drug, as well as to various combinations and sequences of low, fixed doses of one drug with higher varying doses of the other, cultured for two more days in drug-free medium and then assessed for growth response with the sulphorhodamine B assay. Multiple drug effect analysis was used to evaluate the data, which were then related to the amount of DNA damage occurring in the cells which was determined by a fluorescence-enhancement assay for DNA unwinding. Cellular responses were also related to thymidylate synthase topoisomerase I and carboxyl esterase activities, which were assessed by a ligand-binding and a 3H-release assay; a DNA decatenation assay; and a spectrophotometric method, respectively. IC50 values for 5-FU alone in the SW620, HT29 and SNU-C4 cells were 15.3 +/- 0.8, 8.2 +/- 1.3 and 2.2 +/- 0.7 microM, respectively, and for CPT-11 2.0 +/- 0.9, 2.5 +/- 0.5 and 3.8 +/- 0.3 microM, respectively. The differential responses to 5-FU alone were possibly determined by differences in substrate affinity and conversion rate of thymidylate synthase (K(m) of approximately 7.5, 5.0 and 2.5 microM and V0 of approximately 800, 200 and 2400 microM/h, respectively). The comparable cellular responses to CPT-11 alone might be accounted for by the counterbalancing effects of differences in topoisomerase I (1, 1, and 1.5 arbitrary units, respectively) and carboxyl esterase activities (5055 +/- 1789, 4080 +/- 752, 1713 +/- 522 mU/mg, respectively). IC20 CPT-11 prior to 5-FU was additive to synergistic in SW620, HT-29 and SNU-C4 cells (CIs of 0.7 +/- 0.1). By contrast, pre-treatment with IC20 5-FU antagonised the CPT-11-mediated growth inhibition (CIs of 1.9 +/- 0.4, 1.7 +/- 1.1, 2.5 +/- 0.9, respectively). Simultaneous drug treatment did not produce more cell growth inhibition than either drug alone in the SW620 and the HT-29 cells, but was additive or antagonistic in the SNU-C4 cells (CIs of 1.1 +/- 0.3 and 2.2 +/- 1.4), depending on the ratio of the drugs. Increased DNA damage in the SW620 and HT-29 cells was only seen when IC20 CPT-11 preceded IC50 5-FU, resulting in approximately 40 and 25%, respectively, more lesions than for IC50 5-FU alone. In the SNU-C4 cells, not only such a treatment, but also simultaneous drug treatment produced (30 to 60%) more DNA damage than either drug alone. Our results show clear sequence-dependent antiproliferative effects and DNA damage formation by CPT-11 and 5-FU at combinations of low, fixed doses with higher, varying doses in cultured human colon carcinoma cells, and may be of relevance to the design of improved chemotherapeutic regimens in this disease.