qRT-PCR showed that, in stationary phase bacteria, SbrE transcript levels were significantly higher in parent strains relative to their ΔsigB mutants in (i) four strains representing all four L. monocytogenes lineages and (ii) an L. innocua strain (4.8 ± 1.76 to 8.6 ± 0.67 higher log2 SbrE transcript in the parent strain) (Figure 1), supporting σ^B-dependent transcription of SbrE across L. monocytogenes lineages and in L. innocua. While these findings were not necessarily unexpected, they are still valuable as other studies have shown some diversification of the σ^B regulon and variation in σ^B-dependent regulation of conserved genes, among L. monocytogenes lineages and Listeria species [10,37]. qRT-PCR of SbrE transcripts in the L. monocytogenes strain 10403S showed higher transcript levels in early stationary phase cells (OD 1.0 + 3h) as compared to mid-log phase (OD 0.4) or late log phase (OD 1.0) cells, consistent with σ^B-dependent transcription of SbrE (as σ^B is induced in stationary phase cells) [7,11]. While SbrE transcript levels were not induced after exposure of mid-log phase cells to salt stress (Figure S1), they were induced after exposure to oxidative stress (13 mM cumene hydroperoxide [CHP]). SbrE transcript levels were 1.70 ± 0.71 log2 (absolute, non log-transformed fold change of approximately 3) higher in CHP treated cells, relative to mid-log phase cells.

Microarray experiments comparing transcript levels in L. monocytogenes 10403S parent and ΔsbrE strains identified a single gene that showed differential transcript levels (FC ≥ 1.5 and p ≤ 0.05). Specifically, lmo0636 showed 2 fold lower transcript levels in ΔsbrE as compared to the parent strain. lmo0636 transcript levels were also found to be downregulated in ΔsigB, as compared to the parent strain (FC = −2.17; p < 0.05). lmo0636 transcript levels were not found to be significantly different in the microarray comparison between ΔsigB and ΔsbrE. qRT-PCR confirmed lower lmo0636 transcript levels in ΔsbrE as compared to the parent strain (1.38 ± 0.16 log2 lower in ΔsbrE; p < 0.05 one sample t-test). These data indicate that lmo0636 is positively regulated by SbrE.

Proteomics experiments identified two proteins that were differentially expressed (FC ≥ 1.2 and p < 0.05) between the L. monocytogenes parent strain and ΔsbrE. Lmo0637 and Lmo2094 both showed lower protein levels in the ΔsbrE strain (1.45 and 1.2 fold, respectively). Lmo2094 has been annotated as a metal ion binding, class II aldolase/adducin domain protein (Uniprot, www.uniprot.org). Lmo0637, annotated as an UbiE/COQ5 family methyltransferase, is encoded by a gene that forms a 2 gene operon with lmo0636[11]. Hence, the combination of transcriptional and proteomics results indicates that SbrE regulates the expression of the lmo0636-lmo0637 operon. lmo0636 encodes a protein that was annotated as a hypothetical 2Fe-2S cluster/DNA binding protein of the Rrf2 family of regulators, which belongs to the winged helix-turn-helix superfamily of transcriptional regulators [38]. The N-terminal and C-terminal regions of Rrf2 family proteins are generally involved in DNA binding and signaling, respectively, and may function as redox sensors [39]. Interestingly, previous studies were not able to identify lmo0636/lmo0637 transcription patterns that would point towards a specific mechanism for regulation of this operon. While Raengpradub et al. [10] did not find lmo0636 and lmo0637 to be significantly differentially expressed in comparisons of L. monocytogenes 10403S and ΔsigB, in an L. monocytogenes prfA* genetic background (which expressed a constitutively active PrfA), both genes were found to have significantly higher transcript levels in ΔsigB strains, suggesting that they are negatively regulated by σ^B in the presence of an active PrfA [9]. On the other hand, lmo0637 was found to be up-regulated in the host during mouse infection with L. monocytogenes EGD-e, as compared to stationary phase and exponential phase cells grown in BHI, while lmo0636 was reported to be downregulated in the host as compared to stationary phase cells grown in BHI [40]. These data suggest that transcriptional regulation of lmo0636/lmo0637 is highly dependent on environmental conditions and may be fine-tuned by SbrE and σ^B-dependent transcription of sbrE.

As trans-encoded ncRNAs largely act through base pairing with target RNAs, typically the 5′UTR, consequently affecting their translation and/or stability [27], we modeled the putative interaction between SbrE and lmo0636 in silico, using IntaRNA version 1.2.2 [41] (Figure 2). We only found an interaction with a ΔEnergy of −11.75 kcal/mol, indicating limited complementarily between SbrE and lmo0636 (including its 5′UTR). A preliminary target capture experiment that used biotin-labeled SbrE bound to BioMag Streptavidin beads (Qiagen, Valencia, CA, USA) to capture lmo0636 RNA (with subsequent detection by qRT-PCR) also found no evidence for a specific interaction between SbrE and lmo0636. Specifically, levels of lmo0636 RNA recovered were not different from levels of RNA recovered for another L. monocytogenes gene with no evidence for SbrE dependent expression (i.e., lmo0514). Future experiments are thus needed to identify the direct or indirect mechanism by which SbrE may influence the expression of the lmo0636-lmo0637 operon or to identify other SbrE targets.

As Lmo0636 is annotated as a protein that may play a role in oxidative stress response (see Section 2.2), we initially focused on characterizing the oxidative stress survival phenotype of the ΔsbrE mutant constructed here. Initial assays showed relative killing of 1.95 log CFU/mL for the ΔsbrE mutant as compared to 1.11 log CFU/mL for the parent strain after oxidative stress (13 mM CHP) exposure for 15 min, indicating a potentially small but significantly (p = 0.0084) reduced ability to survive oxidative stress for the ΔsbrE mutant (Figure S2).

As the difference in survival between the parent and the ΔsbrE strain was <1 log (i.e., 0.84 ± 0.29 log CFU/mL), follow up experiments were conducted to monitor oxidative stress (13 mM CHP exposure) survival over 60 min. In these experiments, we found no significant differences in log reduction for the parent and the ΔsbrE strain after CHP exposure for 15, 30, and 60 min (p > 0.05), even though the ΔsbrE strain showed numerically higher log CFU reductions, as compared to the parent strain at each time point, with the difference being <1 log at each time point (Table 1 and Figure S3). We thus used a competitive index experiment, which provides a more sensitive approach to identify phenotypic differences between two strains, to compare the oxidative stress resistance between the parent and the ΔsbrE strain. After 13 mM CHP exposure for 15 min the competitive index comparing the Erm^r parent strain and the Erm^s ΔsbrE mutant was 1.63, virtually the same as for the control comparing the Erm^r parent strain to an Erm^s parent (1.50), suggesting no difference in oxidative stress survival between the parent and ΔsbrE strains in this experiment. In all three experiments detailed above we did find evidence for significantly reduced oxidative stress resistance of the ΔsigB strain, including a competitive index of 13.29 for the comparison between the parent strain and the ΔsigB strain. These findings are consistent with previous reports, which showed that σ^B contributes to oxidative stress resistance in L. monocytogenes[8].

Further phenotypic evaluation of the ΔsbrE strain showed no significant effect of the sbrE deletion on (i) ability to survive acid stress (pH 2.5, 1 h; see Table 1 and Figure 3), (ii) ability to survive under salt stress (1.75 M NaCl, 12 h; see Table 1 and Figure S4); (iii) growth under glucose-limiting conditions (0.04% wt/vol glucose, 30 h; see Table 1 and Figure S5), and (iii) growth at 7 °C for 12 days (Table 1 and Figure S6). On the other hand, the ΔsigB strain, which was included as a control, showed (i) significantly higher death rate under acid stress as compared to the parent strain (p = 0.0054) and ΔsbrE (p = 0.0022) (Table 1 and Figure 3); (ii) significantly reduced ability to survive salt stress as compared to the parent strain (p = 0.0039) and ΔsbrE (p = 0.0039) (Table 1 and Figure S4); and (iii) significantly greater increase in cell density under glucose limiting conditions as compared to the parent strain (p = 0.0008) and ΔsbrE (p = 0.0014) (Table 1 and Figure S5). The ΔsigB strain showed a small but significant (p = 0.0371) reduction in growth rate under cold stress compared to wildtype, with a difference of 0.06 ± 0.03 log CFU/mL/day (Table 1 and Figure S6). Susceptibility to infection from the 22 Listeriaphages tested did not differ between wildtype, ΔsbrE, and ΔsigB strains (Table S1).

Our data suggest that SbrE does not contribute to L. monocytogenes survival and growth under a number of stress conditions that are well established to require σ^B for optimal growth and survival. Overall, we found that a SbrE deletion does not affect L. monocytogenes growth under cold stress or energy stress or L. monocytogenes acid stress survival or phage resistance. SbrE did however show a small but significant contribution to the survival of L. monocytogenes 10403S under oxidative stress in our initial experiments; however, this phenotype was not confirmed by subsequent experiments. These data indicate potential contributions of L. monocytogenes SbrE to growth and survival under very specific and defined environmental stress conditions. Interestingly, the characterization of the σ^B-dependent ncRNA SbrA also found no phenotypes for a ΔsbrA mutant strain under the conditions tested [30].

Strains used in this study are listed in Table 2. Stock cultures of all strains were stored at −80 °C in Brain Heart Infusion (BHI) medium containing 15% glycerol. Cultures were streaked onto BHI agar and incubated at 37 °C for 24 h to obtain isolated colonies for inoculation of overnight cultures. Specific growth conditions for each experiment are described below.

A nonpolar internal deletion mutant allele of sbrE was created by splicing by overlap extension (SOE) PCR and allelic mutagenesis, using previously described procedures [44]. Allelic exchange mutagenesis of the wildtype sbrE allele with the mutant allele was confirmed by PCR amplification and direct sequencing of the PCR product (see Table S2 for primers).

qRT-PCR was used to quantify (i) sbrE transcript levels in parent and ΔsigB mutant strains representing the different lineages of L. monocytogenes as well as one L. innocua strain, and (ii) lmo0636 transcript levels in L. monocytogenes parent strain10403S and its isogenic ΔsbrE null mutant. Briefly, cells were grown to stationary phase at 37 °C as previously described [10], with shaking at 230 rpm. After cells reached stationary phase, RNAProtect bacterial reagent (Qiagen, Valencia, CA, USA) was used to stabilize the mRNA according to manufacturer’s instructions. Bacterial cells were collected by centrifugation and stored at −80 °C prior to RNA isolation. RNA extraction was performed using TRI reagent as described previously [45]. Total RNA was incubated with RNasin (Promega, Madison, WI, USA) and RQ1 DNase (Promega) to inhibit RNases and remove DNA contamination, respectively. Further RNA cleanup and concentration was performed using the RNeasy MinElute Cleanup Kit (Qiagen). A NanoDrop ND-1000 spectrophotometer (NanoDrop, Rockland, DE, USA) was used to quantify and assess purity of the RNA. RNA quality and integrity was assessed by the Agilent 2100 Bioanalyzer (Agilent, Santa Clara, CA, USA).

One microgram of RNA from each sample was reverse transcribed to cDNA using random hexamers and reverse transcriptase (TaqMan Reverse Transcription Reagents, Applied Biosystems, Carlsbad, CA, USA) prior to qRT-PCR. To evaluate residual genomic DNA contamination, control reactions without reverse transcriptase were included for each template. qRT-PCR was performed on an ABI Prism 7000 Sequence Detection System (Applied Biosystems), using the TaqMan Universal PCR Master Mix Reagent (Applied Biosystems). Duplicate qRT-PCR reactions were loaded into MicroAmp optical 96-well reaction plates and run using the following program: 1 cycle at 50 °C for 2 min, 1 cycle at 95 °C for 10 min, followed by 40 cycles at 95 °C for 15 s and 60 °C for 1 min. Standard curves for each target template were included to determine the amplification efficiency. All qRT-PCR analyses were performed in triplicate using RNA isolated from three independent biological replicates of cells (see Table S2 for primers and probes). Relative gene transcription levels, i.e., fold changes, were calculated using the efficiency calibrated mathematical model described by Pfaffl [46]. Target transcript levels were normalized to transcript levels of the housekeeping gene rpoB, which displays relatively stable transcript levels under varying experimental conditions [8].

qRT-PCR was used to measure sbrE transcript levels in mid-log phase (OD₆₀₀ 0.4), late log phase (OD₆₀₀ 1.0), early stationary phase (OD₆₀₀ 1.0 + 3 h), and after exposure of mid-log phase cells to either (i) 13 mM cumene hydroperoxide (CHP) (Sigma-Aldrich, St. Louis, MO, USA), 15 min (as described by Oliver et al. [8]), or (ii) 10% NaCl, 15 min. L. monocytogenes 10403S cells were grown as described above. To apply salt stress, an equal volume inoculum of mid-log phase cells was transferred to 5 mL 20% NaCl, and cultures were then incubated at 37 °C with shaking for 15 min. RNA extraction, cDNA synthesis, and qRT-PCR were performed as described above, with the exception that RNA was purified using two phenol-chloroform extractions and one chloroform extraction, followed by RNA precipitation and resuspension in RNase-free water, instead of the RNeasy MinElute Cleanup Kit (Qiagen) procedure described above. Relative gene transcription levels were calculated using the Pfaffl model as described above [46] with target transcript levels normalized to transcript levels of rpoB within samples. Results from all samples were normalized to SbrE transcript levels from a single replicate at mid-log phase (OD 0.4) [47].

L. monocytogenes 10403S parent strain, ΔsbrE, and ΔsigB cells were grown to stationary phase and total RNA was extracted as described above for qRT-PCR analysis. DNA Microarray design and construction were described in a previous study [10]. cDNA synthesis, labeling with dyes, and hybridization were performed as described by Ollinger et al. [9], with the exception that samples were labeled with Cy3 and Cy5 dyes (Amersham Biosciences, Piscataway, NJ, USA). Three replicates using three independent RNA isolations were performed for each microarray comparison (ΔsbrE versus parent strain; ΔsbrE versus ΔsigB). Microarray statistical analysis was performed as described previously [10]. A n-fold change of ≥1.5 was used as the cutoff for the identification of differentially expressed genes.

L. monocytogenes 10403S and ΔsbrE were grown to stationary phase as described above. After growth to stationary phase, bacterial cells from 25 mL of culture were collected by centrifugation. Cell pellets were quick-frozen with liquid nitrogen and stored at −80 °C prior to protein isolation. Proteins were isolated as previously described [48] with slight modifications. Briefly, cell pellets were washed in 1 M triethyl ammonium bicarbonate buffer (pH 8.5) (Sigma-Aldrich, St. Louis, MO, USA) with 0.1% (wt/vol) SDS and 10 μg/mL chloramphenicol (extraction buffer (EB)). Cells were then lysed using a bead beater (BioSpec Mini-Beadbeater-1) in a mixture of 0.5 mm zirconia/silica beads and 1 mL EB. The protein sample was separated from beads by centrifugation and protein concentrations were determined using a noninterfering protein assay kit with bovine serum albumin as the standard (Calbiochem, San Diego, CA, USA) according to the manufacturer’s instructions. 1D SDS-PAGE was used to verify sample protein concentration and quality.

Protein samples were analyzed at the Cornell University Proteomics and Mass Spectrometry Core Facility using shotgun-based quantitative proteomics. A total of 100 μg protein of each sample was denatured, reduced with 5 mM tris-(2-carboxylethyl) phosphine at 37 °C for 1 h and the cysteine residues were blocked with 8 mM methyl methanethiosulfonate for 10 min at room temperature. Protein samples were digested with 10 μg of sequence-grade-modified trypsin at 37 °C for 16 h. Efficiency of protein digestion was assessed by SDS-PAGE. Tryptic peptides from L. monocytogenes parent strain 10403S and ΔsbrE were each labeled with iTRAQ reagents, according to the manufacturer’s protocols (document #4351918A and 4350831C downloaded from docs.appliedbiosystems.com/search.taf; Applied Biosystems). The labeled samples were then combined and fractionated via Isoelectric focusing (IEF) OffGel electrophoresis (OGE) as described below.

The pooled iTRAQ labeled peptides were separated using an Agilent 3100 OFFGEL Fractionator (Agilent, G3100AA, Santa Clara, CA, USA) as described by Yang et al. [49]. Subsequent nanoLC-MS/MS was carried out using a LTQ-Orbitrap Velos (Thermo-Fisher Scientific, San Jose, CA, USA) mass spectrometer equipped with a nano-ion source as reported previously [49], with the Orbitrap Velos operated in positive ion mode with nano spray voltage set at 1.5 kV and source temperature at 175 °C.

All MS and MS/MS raw spectra from iTRAQ experiments were processed using Proteome Discoverer 1.1 (PD1.1, Thermo) for subsequent database search using in-housed license Mascot Daemon (version 2.2.04, Matrix Science: Boston, MA, USA); quantitative processing, protein identification, and data analysis were conducted as described by Yang et al. [49], with some modifications. Briefly, the L. monocytogenes protein sequence database containing 4177 sequence entries downloaded from the Broad Institute (www.broadinstitute.org/annotation/genome/listeria_group/GenomesIndex.html) on 22 May, 2009 was used for database search. The default Mascot search settings included (i) one missed cleavage for full trypsin with fixed MMTS modification of cysteine, (ii) fixed 4-plex iTRAQ modifications on lysine and N-terminal amines, and (iii) variable modifications of methionine oxidation and 4-plex iTRAQ on tyrosine. The peptide mass tolerance and fragment mass tolerance values were 10 ppm and 30 mDa, respectively. To estimate the false discovery rate (FDR), an automatic decoy database search was performed in Mascot. The relative quantitation ratios were normalized (bias-corrected) using the “median ratio” procedure for the iTRAQ 4-plex in each set of experiments. Two biological replicates were analyzed independently.

The Wilcoxon signed rank test was applied to peptide ratios for each identified protein to determine significant changes between strains. The Fisher’s Combined Probability Test was used to combine FDR adjusted Wilcoxon p-values from each replicate into one test statistic for every protein to obtain a combined p-value. Proteins with peptide ratios exhibiting a Fisher’s Combined Probability Test p-value <0.05 and an iTRAQ protein ratio ≥1.2 in both replicates were considered significantly differentially expressed. Statistical analyses were conducted using R Statistical Software (version 2.14.0; The R Foundation for Statistical Computing: Vienna, Austria, 2011).

Acid and oxidative stress survival of ΔsbrE was compared to ΔsigB and the 10403S parent strain. Cells were grown to stationary phase (OD₆₀₀ of 1.0 + 3 h) as described above. For acid stress experiments, 12 N HCl was added to 5 ml aliquot of stationary phase cells to reduce the culture pH to 2.5 as described previously [8]. Bacterial cells were quantified at 10, 30 and 60 min after addition of HCl by plating on BHI agar using a spiral plater (Autoplate 4000; Spiral Biotech, Inc., Norwood, MA, USA). Three independent replicates were performed.

For oxidative stress experiments, 900 μL of stationary phase cells were exposed to 13 mM CHP for 15 min at 37 °C as described previously [8]. Bacterial numbers were quantified by plating as described above. Three independent replicates were performed. In separate experiments, stationary phase cells were also exposed to CHP as described above over a 60 min period, and bacterial numbers were quantified by plating as described above at 15 min, 30 min, and 60 min. At least three independent replicates were performed.

Survival of ΔsbrE and ΔsigB mutants after CHP stress was also examined using competitive-index experiments. Strains used for these experiments included the erythromycin sensitive parent strain 10403S (Erm^S 10403S) and an erythromycin-resistant 10403S derivative (DP-L3903; Erm^R 10403S). Oxidative stress exposure (13 mM CHP) was applied as described above to strains mixed in a 1:1 ratio including (i) Erm^S ΔsbrE and Erm^R 10403S and (ii) Erm^S ΔsigB and the Erm^R 1043S (to evaluate the relative survival of ΔsbrE and ΔsigB compared to 10403S). A control competition experiment was conducted with a 1:1 mixture of Erm^S 10403S and Erm^R 10403S. Competition experiments conducted with and without 13 mM CHP were plated on BHI and incubated at 37 °C for 24 h. A hundred colonies from BHI plates were patched onto BHI agar containing 1 μg erythromycin/ml (BHI-erm). For differential enumeration, total colonies on BHI-erm were subtracted from total colonies on BHI (100). The competitive index was then calculated as the ratio of Erm^R to Erm^S colonies [43,50]. L. monocytogenes 10403S as well as ΔsbrE and ΔsigB strains were also tested for resistance against 22 diverse listeriaphages, using the procedures described by Vongkamjan et al. [51].

Growth of 10403S parent strain, ΔsbrE, and ΔsigB strains under salt, cold, and energy stress conditions was compared. For salt growth experiments, cells were grown to mid-log phase (OD₆₀₀ of 0.4) as described above. A 0.01% inoculum (vol/vol) was transferred to 50 mL pre-warmed BHI broth supplemented with 1.75 M NaCl in a 300 mL nephelo flask (5 μL into 50 mL). Cells were then incubated for 48 h at 37 °C with shaking (230 rpm). Cell numbers were determined, by plating on BHI agar, at specific time points over 48 h.

For cold growth experiments cells were grown to mid-log phase (OD₆₀₀ of 0.4) as described above. A 0.01% inoculum (vol/vol) was transferred to pre-chilled 50 mL BHI broth in a 300 mL nephelo flask. Cells were then incubated at 7 °C for 12 days without shaking, and cell numbers were determined by plating on BHI agar using a spiral plater. For energy stress experiments, carbon starvation was induced by growing cells in defined medium (DM) containing a growth-limiting concentration of glucose (0.04%, wt/vol) [52]. Cells were initially grown in 5 mL of BHI broth at 37 °C overnight with shaking (230 rpm). A 0.1 mL aliquot of the overnight culture was inoculated into 10 mL DM supplemented with 0.4% (wt/vol) glucose and incubated for 12 h with shaking (230 rpm), followed by inoculation of a 0.01 mL aliquot into 10 mL pre-warmed DM containing 0.04% glucose and subsequent incubation for 30 h at 37 °C with aeration. Cell numbers were determined by plating on BHI agar using a spiral plater at specific time points over 30 h. Three biological replicates were performed for each growth experiment.

The Baranyi model [53] was used to estimate maximum growth rates (μ_max) for cold stress experiments, using the NLStools package (version 0.0-5) in R v2.6.2 (The R Foundation for Statistical Computing: Vienna, Austria, 2008). All other statistical analyses were performed with the Statistical Analysis Software (SAS) (version 9.0; SAS Institute, Inc. Cary, NC, USA, 2002). Regression analysis was used to calculate the death rate of cells exposed to pH 2.5, which was expressed as average log CFU death per hour for each strain. Analysis of variance (ANOVA) was used to test if there was significant difference in the death rates between the parent strain (10403S), ΔsbrE, and ΔsigB. ANOVA was also used to test for (i) differences in cell death due to oxidative stress; (ii) differences in growth rate (μmax) of cells exposed to cold stress, which was expressed as increase in cell density in log10 CFU/mL per day; (iii) differences in bacterial numbers after exposure to salt stress and growth under energy stress. Significance was set at p < 0.05 for all statistical analyses.