Promoter Repression by Strong DNA Bending

Abstract

In prokaryotes, genes are typically expressed unless switched off by repressor protein binding at or near the transcriptional start site. The lac operon is a classic model of this genetic switch, which also illustrates repression facilitated by DNA looping. In E. coli, lac repressor (LacI) simultaneously binds two distant operator sequences forming a closed loop containing the regulated promoter. Previous looping studies have shown RNA polymerase (RNAP) and LacI compete for overlapping DNA binding sites. Although elements of the lac operon have been extensively studied, fundamental questions remain. Our current work explores the following questions: Is gene repression by DNA looping observed in the absence of direct competition between LacI and RNAP? To what extent is RNAP excluded from a promoter simply by strong DNA bending? To address these questions we created two families of repression loops where a promoter was placed centrally between two operators. Each family of constructs systematically positioned the promoter on different helical faces of the looped DNA to deconvolute the repressive effect of DNA deformation from topological repression due to steric hindrance. RNAP binding to these constructs was assayed in living E. coli cells (-/+ induction) by a standard enzymatic assay of reporter activity, and by ChIP. Our results support several interesting conclusions: i) DNA bending strain dramatically represses the test promoter (∼100-fold relative to non-looping single-operator constructs) even in the absence of binding competition; ii) DNA looping is destabilized ∼10-fold upon gene induction, emphasizing that inducer-saturated LacI is still competent to bind DNA; iii) promoter topology (i.e. outside vs. inside of the DNA loop) has a ∼10-fold effect on RNAP binding and gene expression. These results illustrate and quantitate how gene regulation can be achieved by DNA deformation in the absence of direct protein-protein binding site competition.