fleQ, the gene encoding the major flagellar regulator of Pseudomonas aeruginosa, is sigma70 dependent and is downregulated by Vfr, a homolog of Escherichia coli cyclic AMP receptor protein

Abstract

The flagellar transcriptional regulator FleQ appears to be the highest-level regulator in the hierarchical regulatory cascade of flagellar biogenesis in Pseudomonas aeruginosa. Except for the posttranslational downregulation of FleQ activity by FleN, an antiactivator, not much is known about the regulation of the fleQ gene or its gene product. Some FleQ homologs in other bacterial species either are positively regulated by another regulator (e.g., CtrA, the master regulator regulating FlbD in Caulobacter crescentus) or are expressed from a sigma70-dependent promoter (e.g., FlgR of Helicobacter pylori). In this study we demonstrated that Vfr, an Escherichia coli CRP homolog known to function as an activator for various genes, including lasR, regA, and toxA, in P. aeruginosa, is capable of repressing fleQ transcription by binding to its consensus sequence in the fleQ promoter. In a DNase I footprint assay, purified Vfr protected the sequence 5'-AATTGACTAATCGTTCACATTTG-3'. When this putative Vfr binding site in the fleQ promoter was mutated, Vfr was unable to bind the fleQ promoter fragment and did not repress fleQ transcription effectively. Primer extension analysis of the fleQ transcript revealed two transcriptional start sites, t1 and t2, that map within the Vfr binding site. A putative -10 region (TAAAAT) for the t2 transcript, with a five-of-six match with the E. coli sigma70 binding consensus, overlaps with one end of the Vfr binding site. A 4-bp mutation and an 8-bp mutation in this -10 region markedly reduced the activity of the fleQ promoter. The same mutations led to the disappearance of the 203-nucleotide fleQ transcript in an in vitro transcription assay. Vfr probably represses fleQ transcription by binding to the Vfr binding site in the fleQ promoter and preventing the sigma factor from binding to the -10 region to initiate transcription.

FIG. 1.

Schematic representation of the fleQ promoter region: 238-bp region including the upstream region of fleQ and the first six codons (boldface type) of its coding region. The deduced amino acid sequence is shown under the codons. The Vfr binding site is enclosed in a box. The putative −10 and −35 sequences for the σ⁷⁰ binding site are underlined. Transcriptional start sites t1 and t2, as determined by primer extension analysis, are shown. The bases mutagenized in the Vfr binding site and the −10 region are italicized and in boldface type, and the designations of the plasmids in which they were generated are indicated. The location of the RT2 primer used for primer extension is indicated by an arrow. The ScaI site is underlined.

FIG. 2.

EMSA of Vfr on the fleQ promoter. (A) An 88-bp radiolabeled fleQ promoter fragment was incubated either with different concentrations of Vfr or with a fixed amount (72.2 nM) of Vfr and different amounts of the unlabeled fleQ promoter fragment as described in Materials and Methods. (B) The mutated fleQ promoter fragment in which CGC was substituted for ACA in the putative Vfr binding site was incubated with different concentrations of Vfr. The amounts of the unlabeled competitor fleQ DNA are indicated above the bars. The concentrations of purified Vfr are indicated below the bars.

FIG. 3.

DNase I footprint of Vfr bound to the fleQ promoter. The fleQ promoter radiolabeled on the coding (A) or antisense (B) strand was incubated with or without purified Vfr and was treated with DNase I as described in Materials and Methods. Lanes G+A, sequencing ladder of the fleQ promoter; lanes +, Vfr present; lanes −, Vfr absent. The region protected from DNase I digestion by Vfr is indicated by a bar and the sequence of the protected region. The DNase I hypersensitive sites are indicated by arrows. The numbers indicate the positions of the nucleotides with respect to the t2 transcriptional start site.

FIG. 4.

Vfr overexpression downregulates FleQ, thereby reducing motility. (A) Motility assay conducted on 0.3% agar plates containing carbenicillin and various IPTG concentrations, as indicated. Spots a and b, PAK harboring the vector control pMMB66EH and the vfr overexpression plasmid pWNP28, respectively. (B and C) Western analysis of the strains used in the motility assay with different IPTG concentrations and with anti-FleQ (B) and anti-Vfr (C) antibodies. Lanes1 to 4 contained lysates from cultures of PAK harboring pMMB66EH grown with 0, 0.01, 0.1, and 1.0 mM IPTG, respectively. Lanes 5 to 8 contained lysates from cultures of PAK harboring pWNP28 grown with 0, 0.01, 0.1, and 1.0 mM IPTG, respectively.

FIG. 5.

Primer extension analysis of fleQ. Thirty micrograms of total RNA isolated from PAK was subjected to primer extension analysis with the RT2 primer and loaded in lane RT2. A sequencing reaction mixture (lanes C, T, A, and G) containing the same primer with pBS-fleQpmutV as the template was loaded and run on a denaturing urea-polyacrylamide gel simultaneously. The sequence of the region spanning the transcriptional start site is shown on the left. The residues marked with an asterisk are the probable start sites determined in this experiment, designated t1 and t2.

FIG. 6.

In vitro transcription analysis of fleQ. DNA templates containing the wild-type fleQ promoter (lane 1) and the 4-bp mutation in its −10 region (lane 2) were analyzed in a transcription reaction by using the E. coli RNA polymerase σ⁷⁰ holoenzyme. ³²P-labeled DNA size markers were electrophoresed in lane M. The sizes of the individual DNA fragments (in bases) are indicated on the left. The arrow indicates the position of the σ⁷⁰-dependent fleQ transcript in lane 1. The asterisk indicates the position of the smaller transcript visible exclusively in lane 2.