We cloned a consensus DNA site for the Escherichia coli FNR protein at different locations upstream of the E. coli melR promoter. FNR can activate transcription initiation at the melR promoter when the FNR binding site is centered around 41, 61, 71, 82, and 92 bp upstream from the transcription start. The SF73 positive control amino acid substitution in FNR interfered with transcription activation by FNR in each case. In contrast, the GA85 positive control substitution reduced activation only at the promoter, where the FNR binding site is 41 bp upstream of the transcript start. The SF73 substitution appears to identify an activating region of FNR that is important for transcription activation at promoters that differ in architecture. Experiments with oriented heterodimers showed that this activating region is functional in the upstream subunit of the FNR dimer at the promoter where FNR binds around 41 bp from the transcript start and in the downstream subunit at the promoters where FNR binds farther upstream.The Escherichia coli FNR and CRP proteins are both global activators of transcription initiation, interacting at a large number of promoters, their activity being triggered by oxygen and glucose starvation, respectively. FNR and CRP possess related primary sequences, and it is presumed that they have homologous structures and evolved from a common origin (for reviews, see references 8 and 10).Target sites for both FNR and CRP span 22 bp, accommodating dimers of both activators. A striking feature of CRPdependent promoters is that there is considerable variation in the location of the CRP binding site from one promoter to another. In contrast, at most naturally occurring FNR-dependent promoters, the 22-bp DNA site for FNR is centered near Ϫ41 (8, 10). Studies in which the same consensus CRP-binding site was cloned at different distances upstream of the same promoter sequence showed that bound CRP could activate transcription when the center of the CRP dimer was located around Ϫ41, Ϫ61, Ϫ71, Ϫ81, or Ϫ91 bp upstream from the start site (7,11). In a parallel study, Bell and coworkers (2) cloned the same DNA site for FNR at different distances upstream of the same core promoter sequence (covering the Ϫ35 region, the Ϫ10 region, and the transcript start). Experiments with the resulting series of promoters showed that FNR was an efficient transcription activator when centered near Ϫ41 but functioned poorly when centered near Ϫ61 and failed to activate transcription when bound farther upstream (2). We reexamined this observation in this work: we found that alteration of the Ϫ35 region of the core promoter sequence permits the construction of a series of semisynthetic FNR-triggered promoters that are active when the DNA site for FNR is centered near Ϫ41, Ϫ61, Ϫ71, Ϫ82, and Ϫ92.By analogy with CRP, it is probable that FNR activates transcription at target promoters by making direct contact with RNA polymerase (4,8,13). The locations of these contact sites in FNR have been mapped by positive control substitutions that ...

The Shigella outer membrane protease IcsP removes the actin assembly protein IcsA from the bacterial surface, and consequently modulates Shigella actin-based motility and cell-to-cell spread. Here, we demonstrate that IcsP expression is undetectable in mutants lacking either of two transcriptional activators, VirF and VirB. In wild-type Shigella spp., virB expression is entirely dependent on VirF; therefore, to circumvent this regulatory cascade, we independently expressed VirF or VirB in Shigella strains lacking both activators and measured both IcsP levels and transcription from the icsP promoter. Our results show that VirB significantly enhanced icsP transcription, even in the absence of VirF. In contrast, when VirF was induced in the absence of VirB, VirF had variable effects. The regulation of icsP is distinctly different from the regulation of the gene encoding its major substrate, icsA, which is activated by VirF and not VirB. We propose that the different pathways regulating icsA and icsP may be critical to the modulation of IcsA-mediated actin-based motility by IcsP.Shigella spp., gram-negative bacterial pathogens cause severe and bloody diarrhea in their human hosts by invading and spreading through the colonic epithelium. Shigella movement within the host cell cytoplasm is dependent on the ability of the bacterium to recruit host cell actin to its surface to form an "actin tail," which propels the bacterium from one cell to another (5,16,29). Actin tail assembly is mediated by a single bacterial protein, IcsA, which is found on the outer surface at one pole of the bacterium (17). This asymmetric localization of IcsA ensures that actin assembly occurs in a directional manner. In its mature form, IcsA is comprised of two domains: the ␣ domain (residues 53 to 758) contains the determinant for actin assembly (14) and extends from the bacterial surface into the extracellular environment, whereas the ␤ domain (residues 759 to 1102) is embedded in the outer membrane (33). The amount of IcsA ␣ domain exposed on the bacterial surface correlates with the efficiency of actin tail formation in the cytoplasm of infected cells (21).IcsP, an outer membrane protease of Shigella, cleaves IcsA between Arg 758 and Arg 759 , removing the entire IcsA ␣ domain from the bacterial surface (8, 13, 15a, 31). Overexpression of IcsP leads to complete removal of the IcsA ␣ domain from the bacterial cell surface (32), whereas genetic disruption of icsP increases the total amount of cell associated IcsA ␣ domain, leading to an increase in the rate of actin-based movement of Shigella (31). Although IcsP is not required for polar localization of IcsA (6,28), it contributes to the maintenance of a tight polar cap of IcsA on the bacterial surface (31). Furthermore, as Shigella enter stationary phase, the amount of cell-associated IcsA ␣ domain decreases dramatically, an effect due at least in part to IcsP (18,32).These data demonstrate that IcsP plays an important role in modulating the amount of the IcsA ␣ domain present on the bacter...

A library of random mutations in the Escherichia coli fnr gene has been screened to identify positive control mutants of FNR that are defective in transcription activation at Class I promoters. Single amino acid substitutions at D43, R72, S73, T118, M120, F181, F186, S187 and F191 identify a surface of FNR that is essential for activation which, presumably, makes contact with the C-terminal domain of the RNA polymerase alpha subunit. This surface is larger than the corresponding activating surface of the related transcription activator, CRP. To identify the contact surface in the C-terminal domain of the RNA polymerase alpha subunit, a library of mutations in the rpoA gene was screened for alpha mutants that interfered with transcription activation at Class I FNR-dependent promoters. Activation was reduced by deletions of the alpha C-terminal domain, by substitutions known to affect DNA binding by alpha, by substitutions at E261 and by substitutions at L300, E302, D305, A308, G315 and R317 that appear to identify contact surfaces of alpha that are likely to make contact with FNR at Class I promoters. Again, this surface differs from the surface used by CRP at Class I CRP-dependent promoters.

The icsP promoter of Shigella spp. is repressed by H-NS and derepressed by VirB. Here, we show that an inverted repeat located between positions ؊1144 and ؊1130 relative to the icsP transcription start site is necessary for VirB-dependent derepression. The atypical location of this cis-acting site is discussed.

Transcriptional silencing and anti-silencing mechanisms modulate bacterial physiology and virulence in many human pathogens. In Shigella species, many virulence plasmid genes are silenced by the histone-like nucleoid structuring protein H-NS and anti-silenced by the virulence gene regulator VirB. Despite the key role that these regulatory proteins play in Shigella virulence, their mechanisms of transcriptional control remain poorly understood. Here, we characterize the regulatory elements and their relative spacing requirements needed for the transcriptional silencing and anti-silencing of icsP, a locus that requires remotely located regulatory elements for both types of transcriptional control. Our findings highlight the flexibility of the regulatory elements' positions with respect to each other, and yet, a molecular roadblock docked between the VirB binding site and the upstream H-NS binding region abolishes transcriptional anti-silencing by VirB, providing insight into transcriptional anti-silencing. Our study also raises the need to re-evaluate the currently proposed VirB binding site. Models of transcriptional silencing and anti-silencing at this genetic locus are presented, and the implications for understanding these regulatory mechanisms in bacteria are discussed.