Sequence specificity in the binding of anti-tumour anthracyclines to DNA: a success of theory.

Up to 1985 apparent contradictions in the results of experimental investigations on the possible sequence specificity in the interaction of the two fundamental antitumour anthracyclines, daunomycin and Adriamycin, with DNA have led to an intellectually confused situation and serious doubts about the existence of any such specificity. Theoretical studies carried out in our laboratory in 1985 demonstrated that these failures were due to the attempt to relate the specificity exclusively to the nature of the two base pairs of the intercalation site. We have shown that a definite specificity can in fact be established but only at the level of triplets of base pairs, comprising in addition to the base pairs of the intercalation site, the neighbouring base pair at the 5' side. Our explicit predictions indicate the preferred triplets G-C, C-G, T-A or G-C, C-G, A-T with the intercalation site between the two GC base pairs, followed relatively closely by the A-T, C-G, T-A triplet with the intercalation site between the AT and CG base pairs. Subsequent extensive experimentation carried out in a number of laboratories confirmed the validity of our concept and the exactitude of our predictions, although some controversy persists among experimentalists as to the nature of the most preferred triplet.