Functional consequences of genome evolution in Listeria monocytogenes: the lmo0423 and lmo0422 genes encode sigmaC and LstR, a lineage II-specific heat shock system

Abstract

Listeria monocytogenes strains belonging to phylogenetic lineage II (serotypes 1/2a, 1/2c, and 3a) carry a lineage-specific genome segment encoding a putative sigma subunit of RNA polymerase (lmo0423, herein referred to as sigC), a gene of unknown function (lmo0422) similar to the padR family of regulators, and a gene that is similar to the rodA-ftsW family of cell wall morphology genes (lmo0421). To understand the function of this set of genes, their expression patterns and the effects of null mutations in the lineage II L. monocytogenes strain 10403S were examined. The data are consistent with the three genes comprising an operon (the sigC operon) that is highly induced by temperature upshift. The operon is transcribed from three different promoters, the proximal of which (P1) depends upon sigC itself. Null mutations in sigC or lmo0422 increase the death rate at lethal temperatures and cause loss of thermal adaptive response, whereas the lmo0421 mutation causes only a loss of the adaptive response component. Only the sigC mutation affects transcription from the P1 promoter, whereas ectopic expression of lmo0422 from the P(SPAC) promoter complements the individual lmo0422 and sigC null mutations, showing that lmo0422 is the actual thermal resistance regulator or effector while sigC provides a mechanism for temperature-dependent transcription of lmo0422 from P1. Our genetic and phylogenetic analyses are consistent with lmo0422-renamed lstR (for lineage-specific thermal regulator)-and sigC comprising a system of thermal resistance that was ancestral to the genus Listeria and was subsequently lost during divergence of the lineage I L. monocytogenes population.

FIG. 1.

Organization and synteny of the RD4 region in genomes of L. monocytogenes and L. innocua strains. Representation of the alignment between the L. monocytogenes strain EGDe, L. innocua, and L. monocytogenes strains H7858 (serotype 4b) and F2365 (serotype 4b) genomes in the RD-4 region is illustrated. Genes are indicated by arrows showing their relative orientations (not drawn to scale). Putative transcription terminators are indicated by a line and filled ellipse. Black arrows represent orthologous genes, stippled arrows indicate nonhomologous genes, and white space indicates absence of the orthologous genes.

FIG. 2.

Northern blot analysis of RNA from the sigC operon. RNA was prepared from L. monocytogenes strain 10403S grown to mid-logarithmic phase at 37°C in brain heart infusion and 30 min after temperature upshift or downshift (48°C or 4°C), addition of NaCl (to 4%), ethanol (to 4%), nisin (to 62.5 U/ml), penicillin G (50 μg/ml), or bile (0.08%). Total RNA was extracted from the treated cells and 50-μg samples of RNA were loaded in the appropriate lanes. The blots are derived from independent experiments with 37°C and 48°C included on both blots as a point of reference. Both blots shown were probed with a 293-bp segment of the sigC gene amplified using the Lm423F1 and Lm423R1 primers (the amplicon extends from +121 to +414 of the lmo0423 coding region).

FIG. 3.

Survival of sigC operon mutants at 60°C. Overnight cultures of the parental strain 10403S and mutant derivatives Δlmo0421, Δlmo0422, and ΔsigC were grown in BHI and inoculated 1:200 into fresh BHI and grown to mid-log phase (OD₆₀₀, ∼0.4) at 37°C. Each mid-log phase culture was then divided in half, with one-half remaining at 37°C and the other half transferred to a 48°C water bath (preadaptation) for 20 min. After the 20-min period, both halves were then shifted to a 60°C water bath. Immediately before the shift to 60°C (time zero) and at intervals after the shift, samples from the cultures were withdrawn, serially diluted, and enumerated by plating onto BHI agar. The results shown here are representative of three independent experiments. Symbols: ▪, 10403S; □, 10403S preadapted; ▴, Δlmo0421; ▵, Δlmo0421 preadapted; •, Δlmo0422; ○, Δlmo0422 preadapted; ⧫, ΔsigC; ⋄, ΔsigC preadapted.

FIG. 4.

Northern blot analysis of sigC operon RNA from the wild type and Δlmo0421, Δlmo0422, and ΔsigC mutants. Overnight cultures of the wild-type 10403S and derived Δlmo0421, Δlmo0422, and ΔsigC mutants were diluted 1:200 into fresh BHI, grown to mid-log phase (OD₆₀₀, ∼0.4), and shifted to 48°C. RNA was extracted from samples harvested before (37) and 30 min after (48) the shift, and 50 μg of RNA was electrophoresed in formaldehyde-agarose gels and transferred to nylon membranes. The membranes were probed with labeled PCR products derived from internal segments of the sigC (+121 to +414 of the lmo0423 coding region) or lmo0421 (+735 to +1191 of the lmo0421 coding region) genes. The observed hybridizing transcripts are shorter in length in the mutant strains due to the length of the individual deletions (sigC deletion = 224 bp; lmo0422 deletion = 219 bp; lmo0421 deletion = 963 bp). To confirm equal loading of RNA, the blots were stripped and reprobed with a probe from a temperature-independent housekeeping gene, lmo0265 (dapE) gene (extending from +190 to +319), which encodes succinyl-diaminopimelate desuccinylase (39).

FIG. 5.

S1 nuclease protection mapping of transcripts from the sigC operon. (A to C) A single-stranded S1 nuclease probe was generated by elongation of the CEX 1 primer (+172 to +151 of sigC) that had been end labeled with [γ-³²P]ATP using T4 kinase. The primer was extended on a single-strand template of an M13 mp19 clone carrying the promoter region using Sequenase. The extension products were digested with HindIII at position −316 relative to the transcription start site, and the end-labeled single-stranded probe molecules (extending from +172 to −316 relative to the sigC start codon) were purified by electrophoresis in a denaturing gel. The purified probe was then mixed with 50 μg of RNA from wild-type or mutant cells, heated to 95°C, annealed at 65°C, and treated with S1 nuclease as previously described (35). The digestion products were dissolved in loading buffer and electrophoresed alongside a sequencing ladder primed with the CEX1 primer. (A) RNA was derived from mid-logarithmic-phase wild-type cells that had been shifted from 37°C to 48°C for 20 min. The positions of the three prominent bands relative to the start codon of the sigC gene are indicated to the right. (B) S1 nuclease reactions were performed on RNA extracted from mid-logarithmic-phase cells of the wild-type and ΔsigC strains that had been shifted from 37°C to 48°C for 20 min. The lengths of the three transcripts are indicated to the left of the image. (C) RNA was extracted from cultures of mid-logarithmic-phase wild-type, Δlmo0421, Δlmo0422, and ΔsigC cells at 37°C (37) and 20 min after upshift to 48°C (48). Only the P1 transcript is shown. (D) Relative map of the operon with the three transcription start sites indicated upstream of the sigC gene. The map is not to scale. (E) Putative promoter sequences upstream of the transcription start sites. Putative −10 and −35 region sequences are underlined.

FIG. 6.

Dot blot analysis of lmo0422 transcripts from P_SPAC-422E strains. Cultures of the Δlmo0422/P_SPAC-422E and ΔsigC/P_SPAC-422E were grown at 37°C to mid-logarithmic phase (OD₆₀₀, ∼0.4), and RNA was extracted from cells harvested prior to or 30 min after addition of IPTG to a final concentration of 1 mM. As a control, RNA was also extracted from mid-logarithmic-phase cells of the Δlmo0422 strain 30 min after upshift of the culture to 48°C. Samples of 20 μg of RNA (each) were then spotted onto nylon, cross-linked by UV light, and then hybridized with a ³²P-labeled PCR product derived from the lmo0422 coding region, which includes 180 bp of the 3′ end of sigC, the entire lmo0422 coding region, and 182 bp of the 5′ end of lmo0421.

FIG. 7.

Ectopic expression of lmo0422 rescues the Δlmo0422 and ΔsigC thermal sensitivity phenotypes. Cells of the wild-type, Δlmo0422, ΔsigC, Δlmo0422/P_SPAC-422E, and ΔsigC/P_SPAC-422E strains were grown overnight in BHI, diluted 1:200 in fresh BHI, and grown to mid-logarithmic phase (OD₆₀₀, ∼0.4). The cultures were then subdivided into several equal portions, and IPTG was then added at 0, 0.5 μM, 1 μM, 2.5 μM, and 5 μM concentrations to the cultures of the P_SPAC-lmo0422E strains. After 30 min, samples of the cultures were removed and enumerated by serial dilution and plating. The remaining portions of the cultures were then transferred to 60°C. Samples were removed at various times after temperature upshift and enumerated by serial dilution and plating. (A) ○, 10403S; □, Δlmo0422; ⧫, Δlmo0422/P_SPAC-422E; ▪, Δlmo0422/P_SPAC-422E plus 0.5 μM IPTG; ▴, Δlmo0422/P_SPAC-422E plus 1 μM IPTG; •, Δlmo0422/P_SPAC-422E plus 2.5 μM IPTG. (B) ○, 10403S; □, ΔsigC; ⧫, ΔsigC/P_SPAC-422E; ▴, ΔsigC/P_SPAC-422E plus 1 μM IPTG; •, ΔsigC/P_SPAC-422E plus 2.5 μM IPTG. Data shown are representative of three or more independent experiments.

FIG. 8.

Multiple hydropathy alignment of RodA-FtsW-like proteins from L. monocytogenes and related species. Proteins belonging to the RodA-FtsW family were identified from the genomes of L. monocytogenes strain EGDe (Lmo), Bacillus subtilis 168 (BSU), Bacillus cereus ATCC10987 (BC), Bacillus anthracis Sterne (BAX), Bacillus thuringiensis serovar konkukian strain 97-27 (BT), Bacillus halodurans C-125 (BH), and Escherichia coli MG1655 (ECO). Multiple alignment was performed using CLUSTAL W (57) using a Blosum scoring matrix, and multiple hydrophobicity plots based on the alignment were generated based on the methods of Hopp and Woods (29, 37) as implemented in the DNAman package. The plot shown was developed using a window size of eight amino acid residues. Each colored plot and the corresponding protein are shown to the right of the plot. Circles in the plot indicate gaps in the alignment or start-stop positions.

FIG. 9.

Phylogenetic analysis of Lmo0421 and other RodA-FtsW-like proteins from L. monocytogenes and related species. Proteins belonging to the RodA-FtsW family were identified from the genomes of L. monocytogenes strain EGDe (Lmo), Bacillus subtilis 168 (BSU), Bacillus cereus ATCC10987 (BC), Bacillus anthracis Sterne (BAX), Bacillus thuringiensis serovar konkukian strain 97-27 (BT), and Bacillus halodurans C-125 (BH) using BLAST analyses. The set of proteins was then subjected to bootstrap analysis (20) using a neighbor-joining search (49). Significant bootstrap values (>50%) from 10,000 repetitions are indicated on the branches of the relevant nodes. Proteins from the genome of each species are filled with the same color. Nodes corresponding to the B. subtilis RodA (purple), B. subtilis SpoVE (green), and the L. monocytogenes Lmo0421 proteins (red) are colored to highlight putatively functionally similar proteins.