Specificity and flexibility of the recognition of DNA helical structure by eukaryotic topoisomerase I

Many studies have been carried out to map the putative binding sites of eukaryotic topoisomerase I on double-stranded DNA. As assayed by the SDS-induced cleavage reaction, results from these studies showed little sequence specificity surrounding the enzyme binding sites. In order to investigate the possible involvement of local helix variations in the recognition of double-stranded DNA by topoisomerase I, we have applied the Calladine-Dickerson rules to analyze the structural variations surrounding over 100 HeLa topoisomerase I cleavage sites on human DNA. Our data suggest that (5′-NRRYRNN-3′/3′-NYYRYNN-5′) and (5′-YRRRYYN-3′/3′-RYYYRRN-5′), in which R is a purine, Y is a pyrimidine and N is any nucleotide, form the consensus recognition sequences for the enzyme. The specific structural features of these two consensus sequences recognized by HeLa topoisomerase I appear to be the local helical twist angle variations. The same consensus sequences are present in the vicinities of other eukaryotic topoisomerase I binding sites. These results have led to a model in which the eukaryotic topoisomerase I enzymes recognize sequence-dependent structural variations of DNA double helices in a specific but flexible mode.