Quantification of topoisomerase-DNA complexes in leukemia cells from patients undergoing therapy with a topoisomerase-directed agent

Several clinically important drugs utilized in cancer chemotherapy inhibit type I (Topotecan) or type II (amsacrine, etoposide) DNA topoisomerases by stabilizing the formation of DNA-topoisomerase complexes (topoisomerase-DNA cross-links). In various cell lines, the magnitude of drug-induced DNA-protein cross-link production correlates with the magnitude of cytotoxicity induced by the drugs. We developed a simple filter-binding assay that can measure drug-induced DNA-protein crosslinks in leukemia cells obtained directly from patients because the assays most widely used for assessment of drug-induced DNA-protein cross-links in cells [sodium dodecyl sulfate (SDS)/KCl precipitation and alkaline elution] are not readily applicable for use on patient material. HL-60 human leukemia cells or freshly isolated patients' leukemia cells were incubated with Topotecan, etoposide, or amsacrine; lysed with SDS; and applied to nitrocellulose filters in a low-salt buffer. DNA is retained on the filter only if it is covalently bound to protein. The amount of DNA retained on the filter is quantified by hibridization to the alu sequence of DNA, which is distributed ubiquitously in the human genome. Using radiolabeled cells, we compared the filter-binding assay directly with the SDS/KCl precipitation assay in the detection of etoposide-or amsacrine-induced DNA-protein cross-links in HL-60 cells and amsacrine-resistant HL-60/AMSA cells. Both the SDS/KCl precipitation assay and the filter-binding assay detected etoposide-induced DNA-protein cross-links in HL-60 and HL-60/AMSA cells and detected a greater frequency of amsacrine-induced DNA-protein cross-links in HL-60 cells than in HL-60/AMSA cells. The filter-binding assay detected DNA-protein cross-links in freshly isolated leukemia cells exposed to Topotecan in vitro. The ratios of DNA retention for Topotecan-treated versus untreated cells from leukemia patients ranged from 1.8 to 11.5. The heterogeneity of this detected cross-linking was as might be expected if the assay were predictive of the antileukemic action of Topotecan, which is variable. This new filter-binding technique may be useful for predicting the sensitivity of individual patients' tumors to drugs that inhibit type I or type II DNA topoisomerases.