Mono-conjugation of an anthraquinone nucleus with a range of naturally occurring amino acids chemically modified at their C-terminus has been adopted as a synthetic approach in the rational design of novel topoisomerase (topo) inhibitors. The biochemistry of topo I and II inhibition has been investigated for a series of 16 new compounds (NU/ICRF 500-515) from which structure-activity relationships have been investigated. Only three compounds could be demonstrated to bind to DNA: two serine derivatives (NU/ICRFs 500 and 506) and an arginine derivative (NU/ICRF 510). In decatenation and relaxation assays with purified enzyme, several compounds were shown to be potent catalytic inhibitors of topo II (100% inhibition at 5 micrograms/mL (10-15 microM) or less) without stabilizing cleavable complex formation. These included the three DNA binding species (of which NU/ICRF 506 was the most active) and a dihydroxyphenylalanine analogue (NU/ICRF 513). Both NU/ICRFs 500 and 506 were further shown to antagonize DNA cleavage induced by amsacrine. Only NU/ICRF 506 unequivocally inhibited the catalytic activity of topo I without induction of DNA cleavage, and was the only combined topo I and II catalytic inhibitor. One compound, NU/ICRF 505 (tyrosine conjugate), stabilized topo I cleavable complexes without inhibiting the catalytic activity of topo I and II. Modifications to the structure of NU/ICRF 505 revealed that the presence of an unhindered hydroxyl on the tyrosine ring and a more hydrophobic ethyl ester at the amino acid C-terminal were both essential, suggesting a highly specific interaction between drug, enzyme and DNA in the ternary complex. Molecular modelling studies suggested that the observed differences in topo inhibition are a consequence of major conformational alterations brought about by small changes in the amino acid substituent, and confirmed a rigid structural requirement for the induction of topo I cleavage, in addition to a less rigid structural requirement for topo II inhibition. A strong correlation was observed between topo inhibition and in vitro cytotoxicity against the human ovarian cancer cell line A2780 (IC50 range 3.4-11.6 microM), suggesting a mechanism of cell kill, at least in part, involving topo inhibition.