Characterization of the ospZ promoter in Shigella flexneri and its regulation by VirB and H-NS

Abstract

OspZ is an effector protein of the type III secretion system in Shigella spp. that downregulates the human inflammatory response during bacterial infection. The ospZ gene is located on the large virulence plasmid of Shigella. Many genes on this plasmid are transcriptionally repressed by the nucleoid structuring protein H-NS and derepressed by VirB, a DNA-binding protein that displays homology to the plasmid partitioning proteins ParB and SopB. In this study, we characterized the ospZ promoter and investigated its regulation by H-NS and VirB in Shigella flexneri. We show that H-NS represses and VirB partially derepresses the ospZ promoter. H-NS-mediated repression requires sequences located between -731 and -412 relative to the beginning of the ospZ gene. Notably, the VirB-dependent derepression of ospZ requires the same VirB binding sites as are required for the VirB-dependent derepression of the divergent icsP gene. These sites are centered 425 bp upstream of the ospZ gene but over 1 kb upstream of the icsP transcription start site. Although these VirB binding sites lie closer to ospZ than icsP, the VirB-dependent increase in ospZ promoter activity is lower than that observed at the icsP promoter. This indicates that the proximity of VirB binding sites to Shigella promoters does not necessarily correlate with the level of VirB-dependent derepression. These findings have implications for virulence gene regulation in Shigella and other pathogens that control gene expression using mechanisms of transcriptional repression and derepression.

Fig 1

Analysis and characterization of the ospZ promoter. (A) Overview of the intergenic region between icsP and ospZ. VirB binding sites identified by Castellanos et al. (15) are indicated and centered 425 bp and 1,160 bp upstream of the ospZ and icsP translation start sites, respectively. (B) Schematic of the 3′-truncated ospZ promoter fragments found in the pospZ_3′t series. For each construct, the position of the 3′ truncation relative to the ospZ ATG is given in parentheses. The 5′ end of the ospZ gene (denoted by the arrow) is found only in pospZ_3′t(+32) and pospZ_3′t(+11). (C) Activity of the pospZ_3′t series in wild-type S. flexneri. pMIC21 serves as the promoterless (P' less) control. Assays were run in triplicate, and the means and standard deviations are shown.

Fig 2

Identification of the transcription start site of the ospZ gene. (A) Primer extension analysis of ospZ transcripts generated by the wild type (2457T) or the wild type carrying the plasmid pDB05. The primer W153 binds within ospZ, and the primer W160 binds within lacZ. Arrows indicate identified transcription start sites, which map to the same nucleotide position. (B) Sequence of the ospZ promoter region and beginning of the gene. The ospZ transcription start site identified by this work is indicated (bold type and designated TSS), and associated potential −10 and −35 elements are shown (bold and underlined). The gene sequence is italicized, and encoded amino acids are given. The sequence bearing strong sequence similarity to the 3′ end of IS3 (92% identity over 185 nt) is highlighted in gray (GenBank number AL391753.1 [32]).

Fig 3

Activity of the PospZ-lacZ fusions in wild-type Shigella and a virB mutant. (A) β-Galactosidase activity resulting from cells carrying either the wild-type full-length promoter (pospZ-5′t(−1613), the full-length promoter containing mutated VirB binding sites (pDB15), or a promoterless control (pMAP07) grown at 37°C. Assays were run in triplicate, and the means and standard deviations are shown.*, P < 0.01. (B) Fold increase in ospZ promoter activity in cells grown with (pBAD-virB) or without (pBAD18) induction of virB. In each case, β-galactosidase activity from the promoterless control in each strain background was subtracted from those generated by cells carrying PospZ-lacZ fusions. The resulting data were normalized to activities obtained in the virB mutant strain carrying pDB15 (construct with the mutated VirB binding sites) when these cells were grown at either 30 or 37°C. Assays were run in triplicate, and the means and standard deviations are shown.

Fig 4

Activity of the ospZ promoter truncation series (pospZ_5′t series) in wild-type E. coli and an hns mutant (A) and wild-type S. flexneri 2457T and a virB mutant derivative (B). pMAP07 serves as the promoterless control. Assays were run in triplicate, and each assay was repeated three times. Means and standard deviations of representative data are shown. Note that the activities of pDB05 and pospZ_5′t(−1613) were found to be identical to each other (data not shown).

Fig 5

Activity of the ospZ promoter (pospZ_5′t(−1613) in a wild-type E. coli strain and an hns mutant derivative after induction of virB from an l-arabinose-inducible plasmid. pMAP07 serves as the promoterless control. Assays were run in triplicate, and the means and standard deviations are shown. Note that the activities of pDB05 and pospZ_5′t(−1613) were found to be identical to each other (data not shown).