Recent evidence casts doubt on the hypotheses that fluoropyrimidine-mediated radiosensitization is related to cytotoxicity or to cell cycle redistribution into the G1/S boundary. We hypothesized that cells that are capable of progressing into S phase in the presence of fluorodeoxyuridine may also be more susceptible to radiation-induced damage. To test this hypothesis, fluorodeoxyuridine (FdUrd)-treated HT29 human colon cancer cells were separated by centrifugal elutriation into four fractions (1-4) containing a range of cells from those at the G1/S boundary (fraction 1) to those which had progressed approximately 11% into S phase (fraction 4). We found that fraction 4 cells showed significantly greater radiosensitization than fraction 1 cells. We also compared the effects of fluorodeoxyuridine on HT29 and SW620 human colon cancer cells. We found that, in contrast with HT29 cells, SW620 cells arrested at the G1/S boundary and were minimally radiosensitized. Finally, we found that an increase in sensitivity was correlated with a decrease in the rate of repair of DNA double-strand and single-strand breaks (assessed by asymmetric field inversion gel electrophoresis and alkaline elution respectively). These findings are consistent with the hypothesis that fluorodeoxyuridine-mediated radiosensitization depends on S phase progression and a decreased ability to repair radiation-induced DNA damage.

The relationship between cell cycle progression and induction of DNA double-strand breaks and cytotoxicity by exposure to fluorodeoxyuridine (FdUrd) was studied in HT29 human colon cancer cells. Fractionation of drug-treated populations by centrifugal elutriation yielded subpopulations having widely divergent abilities to progress through S phase in the presence of the drug. One of these subpopulations, which appeared to undergo coordinated growth arrest, was resistant to FdUrd cytotoxicity and DNA damage. In contrast, the subpopulation which was able to progress furthest through S phase in the presence of FdUrd underwent unbalanced growth arrest (i.e., increase in size and mass out of proportion to DNA synthesis), and displayed both DNA double-strand break formation (assayed by pulsed field gel electrophoresis) and loss of clonogenicity. When cells were elutriated prior to drug treatment, producing fractions enriched in cells at various cell cycle stages, no significant differences in sensitivity to FdUrd-induced cytotoxicity were detected among elutriation fractions. These findings support the model that, in HT29 cells, progression into and through S phase during drug treatment is an important determinant of FdUrd-induced DNA damage and cytotoxicity, but that the cell cycle position at the start of drug exposure is not a critical factor for these effects.