IraL is an RssB anti-adaptor that stabilizes RpoS during logarithmic phase growth in Escherichia coli and Shigella

Abstract

RpoS (σ(S)), the general stress response sigma factor, directs the expression of genes under a variety of stressful conditions. Control of the cellular σ(S) concentration is critical for appropriately scaled σ(S)-dependent gene expression. One way to maintain appropriate levels of σ(S) is to regulate its stability. Indeed, σ(S) degradation is catalyzed by the ClpXP protease and the recognition of σ(S) by ClpXP depends on the adaptor protein RssB. Three anti-adaptors (IraD, IraM, and IraP) exist in Escherichia coli K-12; each interacts with RssB and inhibits RssB activity under different stress conditions, thereby stabilizing σ(S). Unlike K-12, some E. coli isolates, including uropathogenic E. coli strain CFT073, show comparable cellular levels of σ(S) during the logarithmic and stationary growth phases, suggesting that there are differences in the regulation of σ(S) levels among E. coli strains. Here, we describe IraL, an RssB anti-adaptor that stabilizes σ(S) during logarithmic phase growth in CFT073 and other E. coli and Shigella strains. By immunoblot analyses, we show that IraL affects the levels and stability of σ(S) during logarithmic phase growth. By computational and PCR-based analyses, we reveal that iraL is found in many E. coli pathotypes but not in laboratory-adapted strains. Finally, by bacterial two-hybrid and copurification analyses, we demonstrate that IraL interacts with RssB by a mechanism distinct from that used by other characterized anti-adaptors. We introduce a fourth RssB anti-adaptor found in E. coli species and suggest that differences in the regulation of σ(S) levels may contribute to host and niche specificity in pathogenic and nonpathogenic E. coli strains.

Importance: Bacteria must cope with a variety of environmental conditions in order to survive. RpoS (σ(S)), the general stress response sigma factor, directs the expression of many genes under stressful conditions in both pathogenic and nonpathogenic Escherichia coli strains. The regulation of σ(S) levels and activity allows appropriately scaled σ(S)-dependent gene expression. Here, we describe IraL, an RssB anti-adaptor that, unlike previously described anti-adaptors, stabilizes σ(S) during the logarithmic growth phase in the absence of additional stress. We also demonstrate that iraL is found in a large number of E. coli and Shigella isolates. These data suggest that strains containing iraL are able to initiate σ(S)-dependent gene expression under conditions under which strains without iraL cannot. Therefore, IraL-mediated σ(S) stabilization may contribute to host and niche specificity in E. coli.

FIG 1

iraL shares sequence identity with iraM from E. coli K-12 but differs in genetic context. (A) Visualization of regions of the K-12 and CFT073 chromosomes containing iraM and iraL, respectively. Block arrows represent annotated open reading frames from these two strains, and the number below each region indicates the chromosomal location. The region immediately upstream and containing the iraL start codon is detailed. The TSS of iraL (indicated by +1 and a leftward-pointing arrow) was determined by 5′ RACE. The predicted −10 and −35 sites are in bold, and the iraL start codon is underlined. (B) Nucleotide sequence alignment of the coding regions of iraL and iraM. Alignment was carried out with the ClustalW feature and the default parameters in MacVector 9.0.2. It was determined that these genes share 67% nucleotide sequence identity. Stars below the alignment denote regions of sequence identity. (C) Alignment of the amino acid sequences of IraL and IraM. Alignment was carried out with ClustalW as described above. It was determined that these polypeptides share 58.9% amino acid sequence identity. Stars below the alignment denote regions of sequence identity, while periods represent regions of weak similarity.

FIG 2

Distribution of iraL and iraM among selected E. coli pathotypes. The coding sequences of iraL and iraM were queried against complete bacterial genomes in the NCBI database (http://blast.ncbi.nlm.nih.gov/Blast.cgi), and several unsequenced strain collections were subjected to a PCR-based assay for iraL and iraM. Pathotypes with eight or more representatives are shown, with strains represented more than once consolidated into one entry. For the names and sources of all of the strains subjected to these analyses, including pathotypes with fewer than eight representatives, see Tables S3 and S4 in the supplemental material.

FIG 3

IraL affects the levels and stability of σ^S during log phase growth. (A) To assess the effects of IraL on σ^S levels, CFT073, CFT073 ΔiraL, S. sonnei, S. sonnei ΔiraL, K-12/pACYC184 (vector only), and K-12/pIraL (pACYC with iraL under the control of its native promoter) were grown to the log phase (LOG) or the stationary phase (STAT) and immunoblot assays for σ^S were carried out with 10 µg of total soluble protein from these samples. For accompanying Coomassie blue-stained gels that illustrate consistent loading of the protein samples, see Fig. S4 in the supplemental material. Marker (M) lanes contain the Precision Plus Protein Dual Color Standard that cross-reacts with the detection reagents used. (B) To assess the effects of IraL on σ^S stability, CFT073 and isogenic mutants were grown to the mid-log phase as described for panel A, protein synthesis was stopped with Cm, and samples of total protein were precipitated in 10% TCA at 3-min intervals after Cm treatment. Immunoblot assays were carried out with total protein normalized to the OD₆₀₀, and σ^S levels were measured by densitometry. Data points represent the mean percentages of σ^S remaining relative to those at t = 0, and error bars represent ± the standard error of the mean of three replicates.

FIG 4

IraL interacts with RssB. (A) IraL and RssB copurify. Strain AB054 was cotransformed with plasmids encoding CBP-RssB and His₆-IraL, and cell lysates were prepared as described in Materials and Methods. Cobalt beads were used to precipitate CBP-RssB and its interacting partner(s), and immunoblot assays were carried out with protein and anti-CBP or anti-His₆-peroxidase antiserum and visualized as described in Materials and Methods. (B) IraL interacts with full-length RssB and subdomains of RssB, as visualized on MacConkey’s medium. Overnight cultures of BTH101 coexpressing either T18-IraL or T18-IraM fusion protein and the T25-RssB or T25-RssB subdomain were spotted onto MacConkey’s medium plus maltose. Interaction of the proteins of interest is qualitatively shown as red coloration. (C) IraL interacts with full-length RssB and subdomains of RssB, as determined by β-galactosidase assay. Overnight cultures of BTH101 coexpressing either T25-IraL or T25-IraM and the T25-RssB or T25-RssB subdomain were subjected to a β-galactosidase assay as previously described. Bars on the graph represent mean β-galactosidase activities (Miller units [M.U]), and error bars represent the standard error of the mean of three replicates. WT, wild type.