DNA topoisomerase I has been shown to be an important therapeutic target in cancer chemotherapy for the camptothecins as well as for indolocarbazole antibiotics such as BE-13793C and its synthetic derivatives NB-506 and ED-110 [Yoshinari et al. (1993) Cancer Res. 53, 490-494]. To investigate the mechanism of topoisomerase I inhibition by indolocarbazoles, we have studied the induction of DNA cleavage by purified mammalian topoisomerase I mediated by the antitumor antibiotic rebeccamycin and a series of 20 indolocarbazole derivatives. The compounds tested bear (i) various functional groups on the non-indolic moiety (X = CO, CH2, CHOH), (ii) a hydrogen or a chlorine atom at positions 1 and 11 (R2), and (iii) different substituents on the maleimido function (R1 = H, OH, NH2, NHCHO). Half of the ligands have the same carbohydrate moiety as rebeccamycin whereas the other ligands have no sugar residue. The inhibitory potency of the test compounds was assessed in vitro by comparing the cleavage of [32P]-labeled restriction fragments by the enzyme in the absence and presence of the drug. In addition, the DNA-binding properties of these compounds were investigated by means of complementary spectroscopic techniques including electric linear dichroism, and the DNA sequence selectivity was probed by DNase I footprinting. The study shows that the sugar residue attached to the indolocarbazole chromophore is critical for the drug ability to interfere with topoisomerase I as well as for the formation of intercalation complexes. Structure-activity relationships indicate that the presence of chlorine atoms significantly reduces the effects on topoisomerase I whereas the substituents on the maleimido function and the functional group on the non-indolic moiety can be varied without reduction of activity. The results suggest that the inhibition of topoisomerase I by indolocarbazoles arises in part from their ability to interact with DNA. Analysis of the base preferences around topoisomerase I cleavage sites in various restriction fragments indicated that, in a manner similar to camptothecin, the rebeccamycin analogue R-3 stabilized topoisomerase I preferentially at sites having a T and a G on the 5' and 3' sides of the cleaved bond, respectively. By analogy with models previously proposed for camptothecin and numerous topoisomerase II inhibitors which intercalate into DNA, a stacking model for the interaction between DNA, topoisomerase I and indolocarbazoles is proposed. These findings provide guidance for the development of new topoisomerase I-targeted antitumor indolocarbazole derivatives.