Topological unwinding of strong and weak promoters by RNA polymerase. A comparison between the lac wild-type and the UV5 sites of Escherichia coli

Abstract

Two Escherichia coli control regions have been compared in their ability to be unwound by RNA polymerase during formation of the transcriptionally active ("open") complex: the wild-type control region, consisting of two overlapping binding sites P1 and P2, both weakly transcribed, and an "up" P1 mutant, the strong lac L8UV5 promoter. The final complexes were characterized by their topological unwinding, by DNase I and orthophenanthroline footprints, as well as by methylation of unpaired cytosine residues. At the wild-type control region, the RNA polymerase footprint is weak, and single-strand formation is incomplete and slow. The same signals are strong, complete and quickly established at lac L8UV5; yet the final complexes were found to be equally unwound (by 1.7 turns) in the absence of nucleotide substrates as well as during an abortive initiation cycle. At the lac wild-type region, open complex formation occurs slowly enough to permit the measurement of the extent of a single-stranded region and of topological unwinding during the latency period. Not all the final species are active and unwinding appears to precede, in time, full open-complex formation. At the lac UV5 promoter the same conclusion was reached by a different method involving those changes in the various parameters that characterize open-complex formation monitored by an abortive initiation assay, conducted at increasing levels of template superhelicity. From both approaches we conclude that, at these promoters, the formation of the single-stranded region occurs at the expense of a negative change in linking number, initially stored in a closed intermediate, perhaps as negative writhing. Furthermore, abortive transcription assays indicate that the specific initiation efficiency of the species stored at both promoters, P1 and P2, on the wild-type template is increased as a whole with increasing superhelicity (conversion of inactive species to active ones, increased efficiency of active ones). We conclude that negative supercoiling is not an extra-regulatory element of the lactose system, allowing modulation of expression of the wild-type promoter to the profit of P1. Instead, P2 and P1, in the absence of active catabolite receptor protein (CRP-cAMP), appear to be equally weak and to be equally affected by negative supercoiling in the range of superhelical densities examined. The physiological importance of the P1-P2 competition in the regulation of expression in this region is thus questioned. The major effect of CRP-cAMP stimulation appears to be the direct activation at the P1 promoter.