Sequence determinants for the recognition of the fork junction DNA containing the -10 region of promoter DNA by E. coli RNA polymerase

It has been recently suggested that E. coli RNA polymerase can specifically recognize a fork junction DNA structure, suggesting a possible role for such interaction in promoter DNA melting [Guo, Y., and Gralla, J. D. (1998) Proc. Natl. Acad. Sci. U.S.A. 95, 11655-11660]. We have determined here quantitatively, using a site-specific binding assay, the effects of base substitutions within the conserved -10 hexamer in the context of a short fork junction DNA on binding to RNA polymerase. Adenine at position -11 and thymine at position -7 were found to be critical for sequence-specific recognition of the DNA. The identities of bases at positions -9 and -8 were found to be not important for the binding whereas replacement of bases at positions -12 and -10 had a mild negative effect on the binding affinity. It was found that for the binding of fork DNA to RNA polymerase, specific sequence recognition was more important than specific recognition of fork junction DNA structure. The pattern of relative importance of bases in the -10 region for binding RNA polymerase was generally consistent with the sequence conservation pattern observed in nature where positions -11 and -7 are the most conserved. Binding experiments with a series of adenine analogues at position -11 revealed that the N1 nitrogen of adenine was a critical determinant for the preference of the adenine at this position, suggesting a mechanism for the nucleation of promoter DNA melting initiation in which RNA polymerase destabilizes duplex DNA by directly competing with the thymine of the A-T base pair for hydrogen bonding to the N1 position of the -11 nontemplate strand adenine.     

RNA polymerase alters the mobility of an A-residue crucial to polymerase-induced melting of promoter DNA

Strand separation in promoter DNA induced by Escherichia coli RNA polymerase is likely initiated at a conserved A residue at position -11 of the nontemplate strand. Here we describe the use of fluorescence techniques to study the interaction of RNA polymerase with the -11 base. Forked DNA templates were employed, containing the fluorescent base, 2-aminopurine (2AP), substituted at the -11 position in a single-stranded tail comprising the nucleotides on the nontemplate strand at which base pairing is disrupted in an RNA polymerase-promoter complex. We demonstrate that the presence of 2AP instead of an A at position -11 has no major effect on the accessibility of DNA to DNase I or KMnO(4) in the presence or absence of RNA polymerase, thus justifying the use of templates containing the 2AP substitution in the fluorescence studies. A blue shift of the 2AP fluorescence emission maximum is observed in the presence of RNA polymerase. The results of fluorescence anisotropy decay studies indicate that about 60% of the 2AP residues at -11 are immobilized in an RNA polymerase complex. This value is in good agreement with the fraction of 2AP-substituted templates determined to be in a stable, heparin-resistant complex with RNA polymerase. These results are consistent with the residue at -11 being tightly bound in a hydrophobic pocket of the enzyme.