Characterization of enhancer binding by the Vibrio cholerae flagellar regulatory protein FlrC

Abstract

The human pathogen Vibrio cholerae is a highly motile organism by virtue of a polar flagellum, and motility has been inferred to be an important aspect of virulence. It has previously been demonstrated that the sigma(54)-dependent activator FlrC is necessary for both flagellar synthesis and for enhanced intestinal colonization. In order to characterize FlrC binding, we analyzed two FlrC-dependent promoters, the highly transcribed flaA promoter and the weakly transcribed flgK promoter, utilizing transcriptional lacZ fusions, mobility shift assays, and DNase I footprinting. Promoter fusion studies showed that the smallest fragment with wild-type transcriptional activity for flaAp was from positions -54 to +137 with respect to the start site, and from -63 to +144 for flgKp. Gel mobility shift assays indicated that FlrC binds to a fragment containing the region from positions +24 to +95 in the flaAp, and DNase I footprinting identified a protected region between positions +24 and +85. Mobility shift and DNase I footprinting indicated weak binding of FlrC to a region downstream of the flgKp transcription start site. These results demonstrate a relatively novel sigma(54)-dependent promoter architecture, with the activator FlrC binding downstream of the sigma(54)-dependent transcription start sites. When the FlrC binding site(s) in the flaA promoter was moved a large distance (285 bp) upstream of the transcription start site of either flaAp or flgKp, high levels of FlrC-dependent transcription resulted, indicating that this binding region functions as an enhancer element. In contrast, the relatively weak FlrC binding site(s) in the flgK promoter failed to function as an enhancer element at either promoter, suggesting that FlrC binding strength contributes to enhancer activity. Our results suggest that the differences in FlrC binding to various flagellar promoters results in the differences in transcription levels that mirror the relative requirement for the flagellar components within the flagellum.

FIG. 1.

Identification of the flaAp and flgKp transcription start sites. Primer extension of the flaA (A) and flgK promoters (B) was performed in wild-type, ΔrpoN, and ΔflrC V. cholerae strains. The σ⁵⁴- and FlrC-dependent transcription start sites correspond to the G and A residues for flaAp and flgKp, respectively (denoted by asterisks). These transcription start sites are 326 and 45 bp upstream from the translational start codons of FlaA and FlgK, respectively.

FIG. 2.

flaA and flgK promoter lacZ fusions identify downstream elements important for σ⁵⁴- and FlrC-dependent transcription. Various truncated promoter fragments from flaAp (A) and flgKp (B) were fused to lacZ in the reporter plasmid pRS551. On the left, the fragments are depicted with numbering in relation to the transcription start site. The β-galactosidase activity of these promoter constructs was measured in wild-type, ΔrpoN, and ΔflrC V. cholerae strains, as shown on the right. The results represent assays performed in triplicate with the standard deviations shown. Note that all promoter fragments (with the exception of flaAP +67/+206) contain the predicted σ⁵⁴-holoenzyme binding site.

FIG. 3.

Mobility shift assays indicate stronger binding of FlrC to flaAp than flgKp. Mobility shift assays were performed as described in Materials and Methods with MBP-FlrC and ³²P-labeled DNA fragments. The 5′ and 3′ ends of the fragments are noted in relation to the transcription start site. (A) flaAp −54/+206, −54/+137, +67/+206, and −54/+66 fragments were used in mobility shift assays either in the absence (−) or presence (+) of 1.5 nM MBP-FlrC. Fragments corresponding to flaAp (+24/+85) (B), flaAp (+24/+110) (C), flaAp (+24/+137) (D), and flgKp (−63/+144) (E) were incubated either in the absence (−) or presence (+) of MBP-FlrC at final concentrations of 3.0, 1.5, 0.75, 0.375, 0.187, 0.09, 0.046, and 0.023 nM. (F) The percent shifted species in panels B to E were calculated as described in Materials and Methods.

FIG. 4.

DNase I footprint of FlrC at the flaA promoter. A flaAp fragment (−54/+137) was radiolabeled and subjected to DNase I digestion as described in Materials and Methods; the bottom strand is shown. The start site of transcription is indicated by an arrow, the downstream area protected at the lowest FlrC concentrations is delineated by a thick line at right (+32/+85), the region protected at higher FlrC concentrations is delineated by a thin line at right (−42/+24), and the enhanced cleavage site is marked with an asterisk. FlrC was added to final concentrations of 1.5, 0.75, 0.375, 0.187, 0.09, and 0.046 nM.

FIG. 5.

The flaAp FlrC binding site(s) act as an enhancer element. Various native and engineered promoter fragments from flaAp and flgKp were fused to lacZ in the reporter plasmid pRS551. On the left, the fragments are depicted with numbering in relation to the transcription start site. The first and fourth promoters from the top, respectively, depict the native flaAp and flgKp. The FlrC binding regions of flaAp (positions +11 to +110; hatched areas) were moved upstream of the transcription start sites to −285 for flaAp and to −220 for flgKp (the third and eighth promoters from the top, respectively). The FlrC binding regions of flgKp (positions +11 to +144; shaded areas) were moved upstream of the transcription start sites, to −285 for flaAp and to −220 for flgKp (the sixth and seventh promoters from the top, respectively). The β-galactosidase activity of these promoter constructs was measured in wild-type, ΔrpoN, and ΔflrC V. cholerae strains as shown on the right. Results represent assays performed in triplicate, with the standard deviations shown.

FIG. 6.

The G residues in flaAp at positions +38 and +78 are important for FlrC binding and transcription. (A) Mobility shift assays of MBP-FlrC (0.016 to 2.0 nM) incubated with the native flaAp fragment (+24/+95) (upper panel) or the same fragment containing the G+38→T and G+78→T changes (lower panel). MBP-FlrC was added to final concentrations of 2.0, 1.0, 0.5, 0.25, 0.125, 0.062, 0.031, and 0.016 nM. (B) The percent shifted species in panel A was calculated as described in Materials and Methods. (C) The reengineered flaAp containing the FlrC binding sites 285 bp upstream of the start site (+11/+110 fused to −285/+10) and the same promoter fragment with the G + 38→T and/or G + 78→T mutations were fused to lacZ in the reporter plasmid pRS551. The β-galactosidase activity of these promoter constructs was measured in wild-type, ΔrpoN, and ΔflrC V. cholerae strains. The results represent assays performed in triplicate, with the standard deviations shown.

FIG. 7.

Schematic representation of the flaA and flgK promoter regions. The transcriptional start site (position +1) and consensus sequence for E-σ⁵⁴ binding are shown for flaAp (A) and flgKp (B). The region of flaAp protected by FlrC, as deduced by DNase I footprinting, is indicated with a thin line for the top strand and thick line for the bottom strand. Horizontal arrows in flaAp indicate the CGGCAA repeats. Mutagenized bases in flaAp (G+38 and G+78) are indicated by small vertical arrows. The asterisk indicates the flaAp base pair demonstrating enhanced cleavage when bound by FlrC during DNase I digestion. Dashed lines in flgKp indicate the regions of homology between flaAp and flgKp, which are also shown in panel C. The initiating translational start codon for FlgK is shown by a small square. (D) Homology between flaAp and the hemolysin genes VC1888 and VCA0219.