{sigma}B-dependent gene induction and expression in Listeria monocytogenes during osmotic and acid stress conditions simulating the intestinal environment

David Sue1, Daniel Fink2, Martin Wiedmann1 and Kathryn J. Boor1

1 Department of Food Science, Cornell University, Ithaca, NY 14853, USA
2 Department of Statistical Science, Cornell University, Ithaca, NY 14853, USA

Correspondence
Kathryn J. Boor
kjb4{at}cornell.edu


   ABSTRACT
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Listeria monocytogenes must overcome a variety of stress conditions in the host digestive tract to cause foodborne infections. The alternative sigma factor {sigma}B, encoded by sigB, is responsible for regulating transcription of several L. monocytogenes virulence and stress-response genes, including genes that contribute to establishment of gastrointestinal infections. A quantitative RT-PCR assay was used to measure mRNA transcript accumulation for the virulence genes inlA and bsh, the stress-response genes opuCA and lmo0669 (encoding a carnitine transporter and an oxidoreductase, respectively) and the housekeeping gene rpoB. Assays were conducted on mid-exponential phase L. monocytogenes cells exposed to conditions reflecting osmotic (0·3 M NaCl) or acid (pH 4·5) conditions typical for the human intestinal lumen. In exponential-phase cells, as well as under osmotic and acid stress, inlA, opuCA and bsh showed significantly lower absolute expression levels in a L. monocytogenes {Delta}sigB null mutant compared to wild-type. A statistical model that normalized target gene expression relative to rpoB showed that accumulation of inlA, opuCA and bsh transcripts was significantly increased in the wild-type strain within 5 min of acid and osmotic stress exposure; lmo0669 transcript accumulation increased significantly only after acid exposure. It was concluded that {sigma}B is essential for rapid induction of the tested stress-response and virulence genes under conditions typically encountered during gastrointestinal passage. As inlA, bsh and opuCA are critical for gastrointestinal infections in animal models, the data also suggest that {sigma}B contributes to the ability of L. monocytogenes to cause foodborne infections.


Abbreviations: qRT-PCR, quantitative RT-PCR


   INTRODUCTION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
In 1999, the US Centers for Disease Control and Prevention estimated that approximately 2500 cases of human listeriosis, including 500 deaths, occur annually in the United States (Mead et al., 1999). Because the majority of human listeriosis cases are associated with consumption of contaminated food (Mead et al., 1999), the ability of Listeria monocytogenes to survive in food-processing environments, in foods, and under conditions encountered during gastrointestinal passage in the host, plays a critical role in disease transmission. L. monocytogenes physiology is robust, as illustrated by its ability to grow at low temperatures (>0 °C) and to survive under a wide range of conditions, including elevated temperatures (up to 45 °C), low pH (>=2·5) and high osmolarity (10–20 % NaCl) (Cole et al., 1990; Farber & Peterkin, 1991; Sleator et al., 2000).

The stress-responsive alternative sigma factors {sigma}S (RpoS) and {sigma}B transcribe genes contributing to bacterial survival under conditions of environmental stress in Gram-negative and in Gram-positive bacteria, respectively. In addition to its role in stress survival (Badger & Miller, 1995; Cheville et al., 1996; Small et al., 1994), {sigma}S also contributes to virulence gene expression in Gram-negative bacteria such as Salmonella and Yersinia (Fang et al., 1992; Iriarte et al., 1995). {sigma}B, encoded by sigB, was initially identified and characterized in Bacillus subtilis (Haldenwang & Losick, 1980; Igo et al., 1987). {sigma}B also has been reported in Listeria (Becker et al., 1998; Wiedmann et al., 1998), Staphylococcus aureus (Gertz et al., 2000), Bacillus anthracis (Fouet et al., 2000) and Bacillus licheniformis (Brody & Price, 1998). The B. subtilis {sigma}B-dependent stress regulon includes over 100 genes which are induced by heat, acid, ethanol or high-osmolarity conditions, or under deprivation of glucose, oxygen or phosphate (Helmann et al., 2001; Petersohn et al., 2001; Price et al., 2001). In L. monocytogenes, expression of some {sigma}B-dependent genes is induced by low temperature, low pH, elevated osmolarity and entry into stationary phase (Becker et al., 1998, 2000; Cetin et al., 2004; Fraser et al., 2003).

As with the stress-responsive {sigma}S in Gram-negative organisms, multiple lines of evidence also link the stress-responsive {sigma}B to virulence in L. monocytogenes. Through microarray analyses, the L. monocytogenes {sigma}B regulon has been shown to consist of at least 55 genes, including several virulence genes (Kazmierczak et al., 2003). In addition, Nadon et al. (2002) identified a {sigma}B-dependent promoter (P2prfA) upstream of prfA, which encodes the positive regulatory factor A protein. PrfA globally regulates the expression of many L. monocytogenes virulence genes (Chakraborty et al., 1992; Milohanic et al., 2003). A recent transcriptome analysis of PrfA-regulated genes identified several stress and virulence genes that are also {sigma}B-dependent, including inlA, lmo0669, opuCA and bsh (Kazmierczak et al., 2003; Milohanic et al., 2003). InlA contributes to L. monocytogenes invasion of human and guinea pig gastrointestinal epithelial cells and therefore to establishment of systemic L. monocytogenes infections (Gaillard et al., 1991; Lecuit et al., 2001). Recent microarray and transcriptional start-site mapping experiments have shown that one of the four inlA promoters is {sigma}B-dependent (Kazmierczak et al., 2003). Multiple animal studies have also shown that {sigma}B-dependent genes, as well as genes that are co-regulated by both {sigma}B and PrfA, are critical for L. monocytogenes virulence. For example, a {Delta}sigB L. monocytogenes strain exhibits a reduced ability to colonize murine liver and spleen following intraperitoneal inoculation (Nadon et al., 2002). The {sigma}B-dependent opuC operon contributes to the ability of L. monocytogenes to colonize the mouse small intestine (Sleator et al., 2001). Dussurget et al. (2002) also showed that an L. monocytogenes strain with a mutation in the {sigma}B-dependent bsh (encoding a bile-salt hydrolase) is impaired in resistance to bile, has reduced fecal carriage after oral infection in guinea pigs, and has reduced liver colonization after intravenous inoculation of mice. These findings suggest that the PrfA and {sigma}B regulatory systems are closely interwoven and may represent integrated regulatory networks critical for regulating virulence and stress-response gene expression in the host environment and under environmental stress conditions. Based on these observations, we hypothesized that L. monocytogenes {sigma}B specifically contributes to induction of virulence and stress-response gene expression under conditions typically encountered during gastrointestinal passage, including elevated osmolarity and reduced pH.

To quantify {sigma}B-dependent L. monocytogenes gene expression patterns under stress conditions typical for the intestinal environment, we developed quantitative RT-PCR (TaqMan) assays to monitor transcript accumulation for the {sigma}B-dependent genes inlA, lmo0669, opuCA and bsh (Kazmierczak et al., 2003) and the housekeeping gene rpoB. While the empty human stomach can be highly acidic, with documented pH measurements as low as 2·0 (Chowduhry et al., 1996), in the small intestine, bacteria encounter a mildly acidic environment with an elevated osmolarity (pH 4·5–6·5; 0·3 M NaCl) (Chowduhry et al., 1996; Davenport, 1982). inlA, opuCA, lmo0669 and bsh were chosen as target genes since each has a confirmed {sigma}B-dependent promoter (Kazmierczak et al., 2003), and they represent a stress-response gene with a possible role in virulence (opuCA), a putative stress-response gene (lmo0669, which encodes an oxidoreductase; Takami et al., 2002) and experimentally defined virulence genes (inlA, bsh). mRNA transcript accumulation for these genes was measured in mid-exponential-phase L. monocytogenes cells exposed to conditions that reflect osmotic (0·3 M NaCl) or acid (pH 4·5) stress conditions typical of those in the lumen of the human intestine (Chowduhry et al., 1996; Davenport, 1982). Our findings show that {sigma}B is responsible for the rapid (<5 min) induction of both stress-response and virulence genes under different stress conditions, including those relevant within the host.


   METHODS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Bacterial strains.
L. monocytogenes wild-type strain 10403S (serotype 1/2a; Bishop & Hinrichs, 1987) and the isogenic non-polar {Delta}sigB null mutant FSL A1-254 (Wiedmann et al., 1998) were used throughout this study.

Growth conditions and cell collection.
Bacterial cells were cultured in brain heart infusion broth (pH 7·4) (BHI; Difco) at 30 °C with shaking (250 r.p.m.), unless stated otherwise. Briefly, BHI broth was inoculated from single isolated colonies of each strain which had been grown overnight on BHI agar. Overnight broth cultures were diluted 1 : 1000 into 10 ml BHI broth and incubated at 30 °C. When these cultures reached OD600=0·4, they were each diluted again (1 : 200) into 120 ml BHI and incubated until each culture reached OD600=0·4, as previously described (Ferreira et al., 2001, 2003).

Salt and acid stress conditions.
To monitor induction of target gene expression during osmotic (0·3 M NaCl in BHI) or acid (pH 4·5 in BHI) stress conditions, 120 ml aliquots of mid-exponential-phase cells (OD600=0·4) were pelleted by centrifugation at 15 344 g for 5 min and resuspended in 30 ml warm (30 °C) BHI. While we appreciate that centrifugation is likely to impose a stress on cells (e.g. due to oxygen limitation), a centrifugation step prior to stress exposure was necessary to allow for standardized exposure to stress conditions. For salt stress exposure, 4 ml of resuspended cells was simultaneously transferred into 12 ml warm (30 °C) BHI (‘unexposed control’) and into 12 ml warm (30 °C) elevated osmolarity BHI (0·372 M NaCl), to yield a final concentration of 0·3 M NaCl. For the acid stress exposure, bacterial cells were diluted into acidified BHI (pH 4·08) to yield a final pH of 4·5. Resuspended cells were incubated at 30 °C with shaking for 0, 5 or 15 min, after which total RNA was isolated from the bacterial cells. We chose to conduct the experiments reported here at 30 °C to assure minimal interference by active PrfA, since prfA has been shown to be transcribed, but not translated, at this temperature (Johansson et al., 2002). ‘No stress' control samples, which had been treated with preparation steps identical to those of the stress-exposed cells, allowed us to control for any non-stress-specific factors that may have contributed to gene induction during sample preparation. Three independent collections for each salt and acid exposure (along with corresponding non-stress controls) were performed for both wild-type and {Delta}sigB mutant strains.

Total RNA isolation.
Total RNA was purified from bacterial cells immediately following stress exposure using the RNAprotect/RNeasy Midi kit (Qiagen). The manufacturer's instructions for ‘enzymic lysis with mechanical disruption’ were followed, with the following alterations. Bacterial cells were pelleted (13 776 g) for 5 min at 30 °C and immediately resuspended in 16 ml RNAProtect reagent (Qiagen). Mechanical bacterial lysis was performed on ice by sonication for three 20 s intervals (output: 20 W) using a Sonicator 3000 (Misonix). DNase treatments were performed using RNase-free DNase (Qiagen) following the manufacturer's instructions, except that two 175 µl DNase treatments were performed on each column for 60 min at 30 °C. With each application of the DNase solution, the solution was briefly centrifuged through the column and reapplied, to ensure uniform saturation of the column membrane. The final RNA pellet was resuspended in 300 µl RNase-free TE buffer (Ambion). Total nucleic acid concentrations and purity were estimated using absorbance readings (260 nm/280 nm) on a NanoDrop ND-1000 spectrophotometer.

TaqMan quantitative RT-PCR (qRT-PCR).
Gene expression was measured for five target genes (inlA, lmo0669, opuCA, bsh and rpoB) using the ABI Prism 7000 Sequence Detection System (Applied Biosystems). Primer Express software was used to design TaqMan primers and probes following the guidelines of the manufacturer (Table 1). TaqMan probes were synthesized by Megabases Inc. with a 6-carboxyfluorescein (6-FAM) reporter dye at the 5' end and QSY7 dark-quencher dye at the 3' end. Each qRT-PCR reaction was performed using the TaqMan One-Step RT-PCR Master Mix Reagents kit (Applied Biosystems) following the manufacturer's instructions. The RT-PCR reaction mixture (50 µl) included 50 ng total RNA, 25 µl 2x TaqMan RT-PCR Mastermix, 12·5 U Multiscribe reverse transcriptase, 900 nM of each primer and 250 nM of the respective TaqMan probe. Duplicate qRT-PCR reactions were loaded into MicroAmp optical 96-well reaction plates and run using the following reaction conditions: 1 cycle at 48 °C for 30 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. To account for possible genomic DNA contamination in each qRT-PCR reaction, ‘reverse transcriptase-negative’ reactions which excluded the Multiscribe enzyme were run in parallel. During the 40 cycles of PCR, the fluorescence in the reaction accumulates and increases exponentially, based on the amount of starting template. For each TaqMan reaction, the critical threshold cycle (Ct) is defined as the PCR cycle in which the amount of fluorescence in the tube increases beyond an established standard threshold ({Delta}Rn=1·1).


View this table:
[in this window]
[in a new window]
 
Table 1. Taqman primer and probe sequences used in this study

 
Genomic DNA standard curves for each of the five target genes were included in each assay to account for differences in the amplification efficiencies of the five target genes and to serve as positive controls. Dilutions of genomic DNA, representing ~6x107, ~6x104 and ~6x101 chromosomal copies, were used as templates. For the standard curve analyses, the reaction mix consisted of 50 ng (~6x107 genome copies), 0·05 ng (~6x104 genome copies) or 0·00005 ng (~6x101 genome copies) of genomic DNA, 12·5 µl 2x TaqMan RT-PCR Mastermix, 900 nM of each primer and 200 nM of the respective TaqMan probe in a total volume of 50 µl. Phenol/chloroform-purified genomic DNA for use in the standard curve was prepared from an overnight BHI culture of L. monocytogenes strain 10403S, as described by Flamm et al. (1984).

For each RT-PCR reaction, genomic DNA standard curves were used to calculate the absolute (uncorrected) cDNA copy numbers, based on the Ct values obtained. In addition, the genomic DNA standard curves were used to calculate contaminating-DNA copy numbers for each qRT-PCR reaction, based on the Ct values obtained for the ‘reverse transcriptase-negative’ reactions. Subsequently, contaminating-DNA copy numbers were subtracted from the (uncorrected) cDNA copy numbers to calculate final cDNA copy numbers for each qRT-PCR reaction. The averaged final cDNA copy numbers for the duplicate qRT-PCR reactions were used for subsequent analyses. These cDNA copy numbers thus describe the mRNA levels for each gene present in each RNA preparation. The term ‘cDNA copy numbers' is used in this paper to reflect the fact that absolute copy numbers were calculated using a DNA standard curve.

Experimental design and statistical analyses.
On a given collection day, {Delta}sigB and wild-type strain cells were exposed, in parallel, to non-stress control conditions and either the acid or osmotic stress treatment. RNA samples from each collection day were stored at –80 °C under ethanol until used for qRT-PCR. On a given assay day, a qRT-PCR assay for all five target genes was performed on RNA samples collected from one strain, under one treatment condition for all exposure time points. Thus, six separate assays were performed (‘assay days’) on RNA samples collected in each of three independent treatments (‘collection days’).

Analysis of variance (ANOVA) models of the cDNA copy numbers were analysed for each of the genes with the following factors: strain ({Delta}sigB and wild-type), stress (non-stress control, acid or osmotic stress) and time (0, 5 or 15 min). In addition to these experimental treatments, we included collection day and assay day as block effects to control for any systematic bias in expression level. Initial data analysis revealed the cDNA copy numbers to be heteroscedastic and strongly skewed, which is common for count data. Consequently, the natural logarithm of the cDNA copy number, ln(cDNA copy number), was taken to correct the skewness and stabilize the variance of the counts to approximate normality. All statistical analyses were performed in R (Ihaka & Gentleman, 1996). Standard regression diagnostics were computed for all models. p values are reported as p<0·05, p<0·01, p<0·001 or p<0·0001.


   RESULTS
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Descriptive analysis of {sigma}B-dependent gene expression in wild-type and {Delta}sigB strains under osmotic and acid stress
Gene-specific TaqMan primer and probe sets were designed to quantify mRNA levels of the L. monocytogenes genes inlA, lmo0669, opuCA, bsh and rpoB. rpoB was chosen as a housekeeping gene to normalize expression values of inlA, lmo0669, opuCA and bsh, as described for previous RT-PCR studies (Milohanic et al., 2003). In preliminary experiments, all five qRT-PCR primer/probe sets were able to reproducibly detect as few as 60 DNA copies.

qRT-PCR (TaqMan) was performed on total RNA isolated from mid-exponential-phase L. monocytogenes wild-type and {Delta}sigB cells following treatment with acid or osmotic stress conditions for 0 (initial), 5 or 15 min. RNA collection for the initial time point involved exposure of bacterial cells to the given stress condition, followed immediately by RNA isolation; thus, cells collected at t=0 were exposed to stress conditions for a brief (<5 min, total) time period. Analyses of expression patterns were initially performed using the previously described strategy of normalizing the absolute cDNA quantity for each target gene to the cDNA quantity of a housekeeping gene (rpoB) determined for the same RNA sample (Graham et al., 2002; Milohanic et al., 2003; Smoot et al., 2001). A graphical representation of the normalized gene expression patterns (Fig. 1) clearly shows that transcript accumulation for all four target genes (inlA, lmo0669, opuCA and bsh) was reduced in the {Delta}sigB strain (Fig. 1c, d), compared to the wild-type strain (Fig. 1a, b), under acid and osmotic stress conditions. While inlA, opuCA and bsh cDNA levels were consistently lower in the {Delta}sigB strain than in the wild-type strain, even under non-stress conditions (OD600=0·4, exponential-phase cells), lmo0669 cDNA levels were too low to show clear differences between the {Delta}sigB and the wild-type strain.



View larger version (17K):
[in this window]
[in a new window]
 
Fig. 1. Descriptive analysis of relative mean gene expression (target gene mRNA level/rpoB mRNA level on y axis) for non-stress control and acid stress (pH 4·5) treatments in (a) wild-type and (c) {Delta}sigB strains, and for the non-stress control and osmotic stress (0·3 M NaCl) treatment in (b) wild-type and (d) {Delta}sigB null mutant strains. Error bars represent 1 SD (n=3). The order of genes from left to right is inlA, lmo0669, opuCA and bsh; values for non-stress-exposed cells are shown first for each gene, followed by values for stress-exposed cells for each gene.

 
Our data also showed increased accumulation of inlA, opuCA and bsh transcripts within 5 min of either acid or osmotic stress exposure in the wild-type strain, compared to the non-stress control treatment at each time point (Fig. 1). Expression of inlA, opuCA and bsh did not increase in the {Delta}sigB strain. Limited accumulation of the lmo0669 transcript was observed under non-stressed and acid-exposed conditions. For example, expression after 15 min of acid exposure was 0·00018 cDNA molecules/rpoB cDNA molecules, compared to 0·00013 cDNA molecules/rpoB cDNA for the non-exposed control cells at the same time point. Expression of lmo0669 was consistently considerably lower than inlA, opuCA or bsh expression under both acid and osmotic stress conditions, as well as in the non-stress-exposed exponential-phase control cells. The low expression levels for lmo0669 are best illustrated by the raw non-rpoB normalized cDNA levels obtained by qRT-PCR. For example, the raw cDNA levels for lmo0669 under osmotic stress averaged 5x104 copies per 50 ng total RNA (average of all time points), whereas mRNA levels for inlA, opuCA and bsh averaged 2x106, 2x106 and 1·5x106 copies per 50 ng total RNA, respectively. Raw non-rpoB-normalized cDNA levels for lmo0669 measured for exponential-phase non-stress-exposed wild-type cells were also often at or below the qRT-PCR detection limit (~60 cDNA copies).

Statistical analysis of expression data
As a first step of a formal statistical analysis of our expression data, we evaluated whether our data were normally distributed. A normal probability plot of duplicate qRT-PCR measurements for absolute cDNA levels for inlA, lmo0669, opuCA, bsh and rpoB indicated that the data were not normally distributed, and a difference plot showed that the variance of the error distribution was not constant. Data were thus transformed by taking the natural log (ln) of the response variable (cDNA levels), and all data analyses described below were performed using ln-transformed data. The normal probability plot and difference plot of the transformed response [ln(cDNA level)] indicated that the assumptions of linear regression were satisfied and that the transformed data were normally distributed.

rpoB expression data
Transcript accumulation for the non-{sigma}B-dependent gene rpoB previously has been used as an internal control to ‘normalize’ {sigma}B-dependent gene expression levels in L. monocytogenes (Milohanic et al., 2003; Sue et al., 2003). In these previous studies, normalization was carried out by dividing the expression level of the gene of interest by the rpoB expression level determined for each respective RNA sample. However, we found that this approach to normalization did not control for any systematic differences in cDNA levels across collections and/or assays. Analysis of our data suggests that rpoB expression levels are significantly affected by all experimental treatments. Analysis of the boxplots of rpoB expression levels (Fig. 2) and corresponding ANOVA analysis of ln(rpoB expression levels) show that time (p<0·01) and stress (p<0·0001) both have a highly significant effect on rpoB expression levels. Strain also affects rpoB expression levels, although at a lower p value (p<0·05). In addition, we found evidence of systematic variation in expression levels between different RNA collection days (p<0·001). Inspection of the rpoB expression box plots (Fig. 2) shows that rpoB expression patterns under acid stress are clearly different from those under osmotic stress or no stress. rpoB expression under acid stress is not only at a lower absolute level (compared to both osmotic and no stress), but also does not show an increase in expression levels over time, compared to the increasing rpoB expression levels over time for cells exposed to osmotic stress or to no stress (Fig. 2). Thus, rpoB cannot be used to control for any systematic variation in expression levels associated with different collection days and/or assay days. Instead, we used a statistical approach to control for systematic variations in expression not under direct experimental control. We have included ‘collection day’ and/or ‘assay day’ as blocking effects in all of the models described below in order to exclude any biases in gene expression level associated with the collection process.



View larger version (22K):
[in this window]
[in a new window]
 
Fig. 2. Box plots for ln-transformed rpoB expression levels under acid (pH 4·5), osmotic (0·3 M NaCl) and non-stress control treatments in the wild-type (top) and {Delta}sigB strains (bottom). Darkened circles represent the median ln(rpoB level) for each strain, treatment and time combination (n=6). The box represents the 25th and 75th quartiles from the dataset; whiskers represent data ranges; open circles represent outlier data points.

 
{sigma}B-dependent expression of inlA, lmo0669, opuCA and bsh
An ANOVA model with stress, strain and time was fitted to explain variation in gene expression between the wild-type and the {Delta}sigB strains. Due to the low cDNA levels observed for lmo0669, a constant was estimated to produce a positive argument under the ln transformation. The profile likelihood estimate of this constant was computed with the logtrans routine (Venables & Ripley, 2002). Inspection of the plots of the ln(cDNA level) or ln(cDNA level+Constant) (for lmo0669) distribution of gene expression levels in wild-type and {Delta}sigB null mutant strains (Fig. 3) shows that the cDNA levels for inlA, opuCA and bsh were higher in the wild-type strain compared to the {Delta}sigB strain; the differences between the two strains were less pronounced for exponential-phase cells than for acid- and osmotic-stress-exposed cells. Differences in cDNA levels between wild-type and {Delta}sigB strains were also less pronounced for lmo0669 (probably due to the low overall expression levels for this gene).



View larger version (29K):
[in this window]
[in a new window]
 
Fig. 3. Plots of (a) bsh, inlA, opuCA and rpoB, and (b) lmo0669 expression levels for the wild-type and {Delta}sigB null mutant strains under non-stress control, acid (pH 4·5) and osmotic stress (0·3 M NaCl) conditions. Data represent ln-transformed expression values and data for all replicates are shown. +, wild-type expression patterns; open circles, {Delta}sigB values. To accommodate a different scale, which is necessary due to low expression levels for this gene, lmo0669 expression values are shown in (b).

 
ANOVA analysis of pooled data for all three time points also confirmed that strain ({Delta}sigB or wild-type) had a highly significant (p<0·001) effect on expression levels after controlling for all other experimental factors. Statistical analyses comparing expression of each gene between the {Delta}sigB and the wild-type strains under each experimental condition used (no stress; pH 4·5; 0·3 M NaCl) also confirmed that inlA, opuCA and bsh expression was significantly higher in the wild-type strain, even when no stress conditions were imposed (p<0·001 for all three genes and all three conditions, pooled data for all three time points). While no significant differences in lmo0669 expression patterns between the {Delta}sigB and the wild-type strains were observed for the non-stress control cells and cells exposed to 0·3 M NaCl (p>0·05), lmo0669 expression differed significantly between the {Delta}sigB mutant and the wild-type exposed to pH 4·5 (p<0·01).

Statistical analyses of absolute gene expression levels at different times after stress exposure
Initial visual inspection of box plots of absolute ln-transformed cDNA levels for opuCA, inlA, bsh and lmo0669 in the wild-type strain at 0, 5 and 15 min after exposure to acid and osmotic stress and after exposure to control (no stress) conditions surprisingly showed a decrease (e.g. inlA under acid-stress conditions) or no apparent change (e.g. opuCA, bsh and lmo0669 under acid-stress conditions) in cDNA levels under most conditions (Fig. 4). On the other hand, under control (no stress) conditions, ln-transformed cDNA levels for opuCA and inlA decreased (Fig. 4), while cDNA levels for bsh and lmo0669 did not show any appreciable change over time (Fig. 4). Preliminary linear regression analysis with the model described above confirmed these visually observed trends. For example, under osmotic stress conditions, expression of inlA was significantly (p<0·001) higher after 5 and 15 min of osmotic stress exposure compared to the non-exposed control, while no statistically significant differences were observed under acid-stress conditions.



View larger version (27K):
[in this window]
[in a new window]
 
Fig. 4. Boxplots of absolute expression values for bsh, inlA, opuCA, rpoB and lmo0669 in the wild-type strain under non-stress control, osmotic (0·3 M NaCl) and acid (pH 4·5) stress conditions. lmo0669 values represent ln(lmo0669 level+Constant). Addition of a constant was necessary due to the low cDNA levels observed for lmo0669; this constant was computed with the logtrans routine (Venables & Ripley, 2002). Darkened circles represent the median ln expression level for each time and treatment combination (n=3). The box represents the 25th and 75th quartiles from the dataset; whiskers represent data ranges; open circles represent outlier data points.

 
Statistical analyses of relative gene expression levels standardized to rpoB expression levels
While our analyses described above showed that rpoB expression levels cannot be used to control for systematic variations in expression levels associated with different collection days and/or assay days, we reasoned that rpoB can serve as a proxy (or indicator) of the overall cellular mRNA expression level for bacterial cells at a given physiological stage. Thus, standardizing expression levels of different target genes by dividing by rpoB expression levels would allow us to analyse changes in expression levels for a target gene relative to expression of a housekeeping gene under a given physiological condition. We thus calculated relative expression levels standardized to rpoB expression levels for all four putative {sigma}B-dependent genes, such as ln(inlA cDNA copies/rpoB cDNA copies), for the two different stress conditions and the ‘no stress' control, and the three different time points. ANOVA analyses with experimental factors ‘stress' and ‘time’ were performed for each of the genes with collection day as a blocking variable to test for relative induction of inlA, opuCA, lmo0669 and bsh after exposure to osmotic or acid stress. Partial F-tests were used to test specific hypotheses about transcript accumulation under experimental treatments. A Bonferroni correction was used to adjust for the significance level ({alpha}=0·05) under all pair-wise multiple comparisons for stress and time; the Bonferroni-corrected p value for significance was p<0·00208 (Table 2). Accumulation of inlA and opuCA transcripts was significantly higher after 5 and 15 min of exposure to osmotic stress conditions compared to the non-stress-exposed control cells (Table 2; Fig. 5). Under the same osmotic conditions, bsh transcript levels were significantly higher than those of the non-stress-exposed control cells after 5 min of stress exposure (p<0·0001), but not after 15 min (Fig. 5). lmo0669 expression did not differ significantly between osmotic-stress-exposed cells and non-exposed cells at any time point (Table 2). In the acid-stress experiments, expression of all four genes (inlA, opuCA, lmo0669 and bsh) was significantly higher after 5 and 15 min of exposure to acid stress compared to the non-stress-exposed control cells (Table 2). bsh and lmo0669 expression was significantly higher immediately following acid-stress exposure (t=0; initial) compared to the non-stress-exposed control cells (Table 2).


View this table:
[in this window]
[in a new window]
 
Table 2. Differences in the relative gene expression levels between L. monocytogenes wild-type cells exposed to acid or osmotic stress and cells exposed to no stress for 0, 5 and 15 min

NS, Not significant: only values of p<0·00208 were considered significant (reflecting a Bonferroni correction for multiple comparisons). For p<0·00208 but >0·001, actual p values are given. RNA collection for the initial time point (t=0) involved exposure of bacterial cells to the given stress condition, followed immediately by cell collection using centrifugation for 5 min and RNA isolation; thus, cells collected at t=0 were still exposed to a stress condition for a short time period (<5 min).

 


View larger version (37K):
[in this window]
[in a new window]
 
Fig. 5. Box plots of relative expression values for bsh, inlA, opuCA and lmo0669 in the wild-type strain under non-stress control, osmotic (0·3 M NaCl) and acid (pH 4·5) stress conditions. Darkened circles represent the median ln(gene of interest cDNA level/rpoB cDNA level) for each time and treatment combination (n=3). The box represents the 25th and 75th quartiles from the dataset; whiskers represent data ranges; open circles represent outlier data points. WT, wild-type.

 

   DISCUSSION
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Studies using L. monocytogenes {Delta}sigB null mutants have clearly established that {sigma}B contributes to bacterial survival under environmental stress conditions, such as acid, oxidative and carbon-starvation stresses (Becker et al., 2000; Ferreira et al., 2001; Wiedmann et al., 1998). Semi-quantitative transcriptional analyses (primer extension analyses and RT-PCR) have also provided some initial evidence of induction of a few {sigma}B-dependent genes under different stress conditions, such as cold stress, osmotic stress and acid stress (Becker et al., 1998; Fraser et al., 2003; Sue et al., 2003). Transcriptional gus reporter fusions have been used to confirm {sigma}B-dependent expression and induction of the opuC operon encoding an osmolyte transporter system (Sue et al., 2003). Recent microarray analysis has identified at least 55 genes belonging to the {sigma}B regulon of L. monocytogenes, including virulence genes and genes associated with heat, osmotic, acid and low temperature related stress responses (Kazmierczak et al., 2003). While these various studies have provided compelling initial data on the contribution of L. monocytogenes {sigma}B to gene expression in this organism, quantitative data on {sigma}B-dependent gene expression and induction patterns under different stress conditions, and {sigma}B-dependent expression of virulence genes have not been available. We thus developed qRT-PCR (TaqMan) primers and probes to allow mRNA quantification of four {sigma}B-dependent genes (inlA, opuCA, lmo0669 and bsh) and a housekeeping control gene (rpoB). Our data reported here provide strong evidence that (i) expression of the virulence genes inlA and bsh and the stress-response genes opuC and lmo0669 is {sigma}B-dependent, confirming our previously reported microarray results (Kazmierczak et al., 2003), (ii) {sigma}B provides a mechanism for rapid (<5 min) induction of gene expression, and (iii) {sigma}B is required for rapid induction of expression of L. monocytogenes genes important for gastrointestinal infection under stress conditions typically encountered by this pathogen in the gastrointestinal tract.

Accurate quantification of bacterial mRNA expression
Accurate collection, analysis and interpretation of qRT-PCR (e.g. using TaqMan or SYBR green technology) data to measure gene expression patterns in eukaryotic and prokaryotic organisms is challenging. The approaches used successfully in eukaryotic systems cannot necessarily be directly applied to prokaryotic systems (Vandecasteele et al., 2001). In particular, use of one or multiple eukaryotic ‘normalizer’ genes, which show constant expression levels under different stress conditions (e.g. the gene encoding {beta}-actin), is a standard approach for normalizing mRNA levels from a target gene relative to an internal standard, permitting researchers to control for differences in cell numbers, mRNA levels and mRNA integrity (Giulietti et al., 2001; Livak & Schmittgen, 2001). By extension, many investigators have used 16S rRNA or bacterial housekeeping genes (e.g. rpoB or gyrA; Graham et al., 2002; Milohanic et al., 2003; Smoot et al., 2001) to normalize target gene expression values to an internal standard to account for experimental differences among samples and treatments. While data generated by these approaches generally appear to correlate well with gene expression levels measured by other approaches, such as microarrays (Graham et al., 2002; Smoot et al., 2001), other investigators have provided convincing evidence to show that the use of an internal RNA standard (i.e. 16S rRNA or housekeeping gene mRNA) for quantifying bacterial gene expression is of questionable value, given the rapid and exponential growth kinetics of bacteria and the resulting significant changes in expression of housekeeping genes that can be observed during in vitro bacterial growth and stress exposure (Hansen et al., 2001; Vandecasteele et al., 2001). Our results described here are consistent with the observations of Vandecasteele et al. (2001), who also showed that expression of housekeeping genes is rapidly upregulated after inoculation of bacteria into fresh BHI media. Specifically, we saw upregulation of rpoB cDNA levels after 5 and 15 min of inoculation into BHI and after inoculation into 0·3 M NaCl BHI, a stress condition that does not prevent growth of L. monocytogenes. On the other hand, absolute rpoB cDNA levels remained unchanged after inoculation into pH 4·5 BHI, a stress condition that inhibits growth of L. monocytogenes (Ferreira et al., 2003). These data illustrate the difficulties of quantifying relative mRNA expression levels in bacteria exposed to different stress conditions. Normalization to initial bacterial cell numbers will not account for differences in bacterial lysis efficiency under different stress conditions (Silva & Batt, 1995) or for RNA losses during purification. Normalization by total RNA amounts is complicated by the fact that bacterial cell size and RNA contents may differ considerably by growth phase and physiological conditions (Hansen et al., 2001). Furthermore, both normalization by bacterial cell numbers and total bacterial RNA, while feasible in pure culture, cannot easily be accomplished with samples collected from more complex environments, for example during in vitro or in vivo host cell infection. Finally, we demonstrated that rpoB cDNA levels cannot be used to control for systematic variation in expression level across different collections. However, these effects can be accounted for by including blocking effects in the ANOVA model. The use of blocking effects can also be extended to the more complex environments mentioned above. While rpoB expression levels cannot be used as an effective internal control, since rpoB is a housekeeping gene, rpoB may be used to normalize target gene expression patterns to allow statistical analyses of changes in target gene expression levels relative to the overall cellular mRNA expression level for bacterial cells under various physiological conditions. This conclusion is supported by the fact that Milohanic et al. (2003) observed a good correlation between macroarray expression data and SYBR Green qRT-PCR expression data normalized to rpoB in L. monocytogenes.

Role of {sigma}B-dependent genes and gene expression in virulence
The qRT-PCR experiments described here not only confirm that inlA, opuCA, lmo0669 and bsh are transcribed in a {sigma}B-dependent manner but, more importantly, provide clear evidence of the importance of {sigma}B for regulating gene expression during osmotic and acid stress conditions typical for those found in the lumen of the human intestine (Chowduhry et al., 1996; Davenport, 1982). These findings are particularly informative for inlA, since expression of this gene has been shown to be regulated by four promoters, and also since no quantitative data on the contributions of these different promoters to inlA expression have been reported previously. Our data clearly indicate that the previously confirmed {sigma}B-dependent inlA P4 promoter (Kazmierczak et al., 2003) may be the major promoter responsible for activating inlA expression during intestinal infection. While Milohanic et al. (2003) and others (Dussurget et al., 2002) have previously shown that inlA, opuCA and bsh are PrfA-regulated genes, our data show that {sigma}B is critical for activating expression of all three of these genes. Interestingly, the PrfA binding box for all three of these genes does not overlap with the –35 region of the confirmed {sigma}B-dependent promoters for any of these genes (Kazmierczak et al., 2003), suggesting that PrfA-dependent expression of these genes may be independent of the {sigma}B-dependent promoter and that it plays a minor role in their expression under the in vitro stress conditions tested here. It is also highly unlikely that the low expression values observed for inlA, opuCA and bsh in the {Delta}sigB null mutant are caused indirectly by reduced prfA expression from the {sigma}B-dependent prfA P2 promoter, since it has previously been shown that loss of {sigma}B alone in a {Delta}sigB null mutant appears to be compensated for (probably by enhanced expression from other prfA promoters), such that the {Delta}sigB null mutant shows no attenuated tissue culture cytopathogenicity in mouse L-cells, which are invaded by an inlA-independent pathway (Freitag & Portnoy, 1994; Nadon et al., 2002). It is important to emphasize that all experiments described here were conducted at 30 °C to assure minimal interference by active PrfA, since prfA has been shown to be transcribed, but not translated, at this temperature (Johansson et al., 2002). Further studies in prfA and sigB null mutants and double mutants grown at different temperatures will thus be required to develop a complete understanding of the interactions between {sigma}B and PrfA in the regulation of various virulence and stress-response genes. Our data shown here, together with data reported by Milohanic et al. (2003) and by Nadon et al. (2002), further add to the increasing evidence that interactions and overlaps between {sigma}B- and PrfA-dependent regulation of gene expression appear to be critical for assuring appropriate gene expression patterns during L. monocytogenes–host interactions.

It is interesting to note that, while inlA, opuCA and bsh all showed fairly high absolute and relative (to rpoB) expression values, lmo0669, the only gene monitored without either a verified or plausible role during intestinal infection, showed extremely low expression values. lmo0669 encodes a putative protein with homology to an oxidoreductase (Takami et al., 2002), and we hypothesized that the protein encoded by lmo0669 might be involved in a {sigma}B-dependent oxidative stress response and thus be important for intravacuolar survival. Interestingly, lmo0669 has also been identified by Milohanic et al. (2003) as a PrfA-dependent gene by microarray experiments, even though it is not preceded by a recognizable PrfA box. Further expression studies on lmo0669 and other PrfA- and {sigma}B-dependent genes with putative or confirmed roles during intracellular infection, but without a role during intestinal infection, will be necessary to further dissect the relative contributions of {sigma}B and PrfA at various (extra- and intracellular) stages of infection. Since it has been clearly shown that mRNA levels do not always accurately predict the amount of protein synthesized from a given gene (e.g. Lee et al., 2003), future experiments including both mRNA and protein expression measurements will be necessary to elucidate the contributions of {sigma}B-dependent gene expression to expressed protein levels.

Conclusions
While microarray studies allow for large-scale gene expression monitoring, more sensitive and quantitatively reliable expression monitoring tools are necessary to accurately measure bacterial gene expression patterns, including those occurring in complex environments (e.g. in host cells) where only small amounts of bacterial RNA, which may also be contaminated with RNA from other organisms, can be obtained. While the use of quantitative RT-PCR to confirm induction trends observed in microarray analyses has been described previously (Graham et al., 2002; Smoot et al., 2001), few studies have applied stringent statistical methods to evaluate qRT-PCR-based expression data. While qRT-PCR has the potential to allow for detection and quantification of subtle changes in gene expression, its use within complex experimental designs necessitates statistical analysis and interpretation of the data. The class of ANOVA models employed in this paper allows for the simultaneous estimation, and therefore control, of many factors. These factors may be manipulated by the researcher or may be outside the researcher's control. Generalized linear models (GLM), an extension of ANOVA, represent an alternative statistical approach with the advantage that they do not require the counts to be normally distributed. In our study, we used ANOVA models to analyse expression data for previously identified {sigma}B-dependent genes using known amounts of pure bacterial RNA collected during in vitro stress-exposure experiments. Inclusion of an internal control housekeeping gene (rpoB) in our analyses allowed us to normalize for different physiological states of bacterial cells as well as for variation in bacterial cell lysis efficiency. For future experiments in which the accurate quantification of bacterial mRNA levels will not be possible, such as the recovery of bacterial RNA from infected host cells, inclusion of additional housekeeping genes will avoid overreliance on a single gene for normalization. The approaches used here to analyse qRT-PCR-based expression data provide an appropriate strategy for statistical analysis of expression data, applicable beyond L. monocytogenes and {sigma}B-dependent gene expression patterns.


   ACKNOWLEDGEMENTS
 
This work was supported in part by the National Institutes of Health Award No. RO1-AI052151-01A1 (to K. J. B.). Development of L. monocytogenes TaqMan primer and probes was supported by a US Department of Agriculture Special Research grant to M. W. (2002-34459-11758). We thank K. Nightingale for help with statistical analyses.


   REFERENCES
TOP
ABSTRACT
INTRODUCTION
METHODS
RESULTS
DISCUSSION
REFERENCES
 
Badger, J. L. & Miller, V. L. (1995). Role of RpoS in survival of Yersinia enterocolitica to a variety of environmental stresses. J Bacteriol 177, 5370–5373.[Abstract]

Becker, L. A., Cetin, M. S., Hutkins, R. W. & Benson, A. K. (1998). Identification of the gene encoding the alternative sigma factor {sigma}B from Listeria monocytogenes and its role in osmotolerance. J Bacteriol 180, 4547–4554.[Abstract/Free Full Text]

Becker, L. A., Evans, S. N., Hutkins, R. W. & Benson, A. K. (2000). Role of {sigma}B in adaptation of Listeria monocytogenes to growth at low temperature. J Bacteriol 182, 7083–7087.[Abstract/Free Full Text]

Bishop, D. K. & Hinrichs, D. J. (1987). Adoptive transfer of immunity to Listeria monocytogenes. The influence of in vitro stimulation on lymphocyte subset requirements. J Immunol 139, 2005–2009.[Abstract/Free Full Text]

Brody, M. S. & Price, C. W. (1998). Bacillus licheniformis sigB operon encoding the general stress transcription factor sigma B. Gene 212, 111–118.[CrossRef][Medline]

Cetin, M. S., Zhang, C., Hutkins, R. W. & Benson, A. K. (2004). Regulation of transcription of compatible solute transporters by the general stress sigma factor, {sigma}B, in Listeria monocytogenes. J Bacteriol 186, 794–802.[Abstract/Free Full Text]

Chakraborty, T., Leimeister-Wachter, M., Domann, E., Hartl, M., Goebel, W., Nichterlein, T. & Notermans, S. (1992). Coordinate regulation of virulence genes in Listeria monocytogenes requires the product of the prfA gene. J Bacteriol 174, 568–574.[Abstract]

Cheville, A. M., Arnold, K. W., Buchrieser, C., Cheng, C. M. & Kaspar, C. W. (1996). rpoS regulation of acid, heat, and salt tolerance in Escherichia coli O157 : H7. Appl Environ Microbiol 62, 1822–1824.[Abstract]

Chowduhry, R., Sahu, G. K. & Das, J. (1996). Stress response in pathogenic bacteria. J Biosci 21, 149–160.

Cole, M. B., Jones, M. V. & Holyoak, C. (1990). The effect of pH, salt concentration and temperature on the survival and growth of Listeria monocytogenes. J Appl Bacteriol 69, 63–72.[Medline]

Davenport, H. W. (1982). Physiology of the Digestive Tract: an Introductory Text, 5th edn. Chicago: Year Book Medical Publishers.

Dussurget, O., Cabanes, D., Dehoux, P., Lecuit, M., Buchrieser, C., Glaser, P. & Cossart, P. (2002). Listeria monocytogenes bile salt hydrolase is a PrfA-regulated virulence factor involved in the intestinal and hepatic phases of listeriosis. Mol Microbiol 45, 1095–1106.[CrossRef][Medline]

Fang, F. C., Libby, S. J., Buchmeier, N. A., Loewen, P. C., Switala, J., Harwood, J. & Guiney, D. G. (1992). The alternative {sigma} factor KatF (RpoS) regulates Salmonella virulence. Proc Natl Acad Sci U S A 89, 11978–11982.[Abstract]

Farber, J. M. & Peterkin, P. I. (1991). Listeria monocytogenes, a food-borne pathogen. Microbiol Rev 55, 476–511.[Medline]

Ferreira, A., O'Byrne, C. P. & Boor, K. J. (2001). Role of {sigma}B in heat, ethanol, acid, and oxidative stress resistance and during carbon starvation in Listeria monocytogenes. Appl Environ Microbiol 67, 4454–4457.[Abstract/Free Full Text]

Ferreira, A., Sue, D., O'Byrne, C. P. & Boor, K. J. (2003). Role of Listeria monocytogenes {sigma}B in survival of lethal acidic conditions and in the acquired acid tolerance response. Appl Environ Microbiol 69, 2692–2698.[Abstract/Free Full Text]

Flamm, R. K., Hinrichs, D. J. & Thomashow, M. F. (1984). Introduction of pAM beta 1 into Listeria monocytogenes by conjugation and homology between native L. monocytogenes plasmids. Infect Immun 44, 157–161.[Medline]

Fouet, A., Namy, O. & Lambert, G. (2000). Characterization of the operon encoding the alternative {sigma}B factor from Bacillus anthracis and its role in virulence. J Bacteriol 182, 5036–5045.[Abstract/Free Full Text]

Fraser, K. R., Sue, D., Wiedmann, M. & Boor, K. J. (2003). Role of {sigma}B in regulating the compatible solute uptake systems of Listeria monocytogenes: osmotic induction of opuC is {sigma}B-dependent. Appl Environ Microbiol 69, 2015–2022.[Abstract/Free Full Text]

Freitag, N. E. & Portnoy, D. A. (1994). Dual promoters of the Listeria monocytogenes prfA transcriptional activator appear essential in vitro but are redundant in vivo. Mol Microbiol 12, 845–853.[Medline]

Gaillard, J. L., Berche, P., Frehel, C., Gouin, E. & Cossart, P. (1991). Entry of L. monocytogenes into cells is mediated by internalin, a repeat protein reminiscent of surface antigens from Gram-positive cocci. Cell 65, 1127–1141.[Medline]

Gertz, S., Engelmann, S., Schmid, R., Ziebandt, A. K., Tischer, K., Scharf, C., Hacker, J. & Hecker, M. (2000). Characterization of the {sigma}B regulon in Staphylococcus aureus. J Bacteriol 182, 6983–6991.[Abstract/Free Full Text]

Giulietti, A., Overbergh, L., Valckx, D., Decallonne, B., Bouillon, R. & Mathieu, C. (2001). An overview of real-time quantitative PCR: applications to quantify cytokine gene expression. Methods 25, 386–401.[CrossRef][Medline]

Graham, M. R., Smoot, L. M., Migliaccio, C. A. & 7 other authors (2002). Virulence control in group A Streptococcus by a two-component gene regulatory system: global expression profiling and in vivo infection modeling. Proc Natl Acad Sci U S A 99, 13855–13860.[Abstract/Free Full Text]

Haldenwang, W. G. & Losick, R. (1980). Novel RNA polymerase {sigma} factor from Bacillus subtilis. Proc Natl Acad Sci U S A 77, 7000–7004.[Abstract]

Hansen, M. C., Nielsen, A. K., Molin, S., Hammer, K. & Kilstrup, M. (2001). Changes in rRNA levels during stress invalidate results from mRNA blotting: fluorescence in situ rRNA hybridization permits renormalization for estimation of cellular mRNA levels. J Bacteriol 183, 4747–4751.[Abstract/Free Full Text]

Helmann, J. D., Wu, M. F., Kobel, P. A., Gamo, F. J., Wilson, M., Morshedi, M. M., Navre, M. & Paddon, C. (2001). Global transcriptional response of Bacillus subtilis to heat shock. J Bacteriol 183, 7318–7328.[Abstract/Free Full Text]

Igo, M., Lampe, M., Ray, C., Schafer, W., Moran, C. P. & Losick, R. (1987). Genetic studies of a secondary RNA polymerase sigma factor in Bacillus subtilis. J Bacteriol 169, 3464–3469.[Medline]

Ihaka, R. & Gentleman, R. (1996). R: a language for data analysis and graphics. J Comp Graph Stat 5, 299–314.

Iriarte, M., Stainier, I. & Cornelis, G. R. (1995). The rpoS gene from Yersinia enterocolitica and its influence on expression of virulence factors. Infect Immun 63, 1840–1847.[Abstract]

Johansson, J., Mandin, P., Renzoni, A., Chiaruttini, C., Springer, M. & Cossart, P. (2002). An RNA thermosensor controls expression of virulence genes in Listeria monocytogenes. Cell 110, 551–561.[Medline]

Kazmierczak, M. J., Mithoe, S. C., Boor, K. J. & Wiedmann, M. (2003). Listeria monocytogenes {sigma}B regulates stress response and virulence functions. J Bacteriol 185, 5722–5734.[Abstract/Free Full Text]

Lecuit, M., Vandormael-Pournin, S., Lefort, J., Huerre, M., Gounon, P., Dupuy, C., Babinet, C. & Cossart, P. (2001). A transgenic model for listeriosis: role of internalin in crossing the intestinal barrier. Science 292, 1722–1725.[Abstract/Free Full Text]

Lee, P. S., Shaw, L. B., Choe, L. H., Mehra, A., Hatzimanikatis, V. & Lee, K. H. (2003). Insights into the relation between mRNA and protein expression patterns: II. Experimental observations in Escherichia coli. Biotechnol Bioeng 84, 834–841.[CrossRef][Medline]

Livak, K. J. & Schmittgen, T. D. (2001). Analysis of relative gene expression data using real-time quantitative PCR and the 2(-Delta Delta C(T)) Method. Methods 25, 402–408.[CrossRef][Medline]

Mead, P. S., Slutsker, L., Dietz, V., McCaig, L. F., Bresee, J. S., Shapiro, C., Griffin, P. M. & Tauxe, R. V. (1999). Food-related illness and death in the United States. Emerg Infect Dis 5, 607–625.[Medline]

Milohanic, E., Glaser, P., Coppee, J. Y., Frangeul, L., Vega, Y., Vazquez-Boland, J. A., Kunst, F., Cossart, P. & Buchrieser, C. (2003). Transcriptome analysis of Listeria monocytogenes identifies three groups of genes differently regulated by PrfA. Mol Microbiol 47, 1613–1625.[CrossRef][Medline]

Nadon, C., Bowen, B., Wiedmann, M. & Boor, K. J. (2002). {sigma}B contributes to PrfA-mediated virulence in Listeria monocytogenes. Infect Immun 70, 3948–3952.[Abstract/Free Full Text]

Petersohn, A., Brigulla, M., Haas, S., Hoheisel, J. D., Volker, U. & Hecker, M. (2001). Global analysis of the general stress response of Bacillus subtilis. J Bacteriol 183, 5617–5631.[Abstract/Free Full Text]

Price, C. W., Fawcett, P., Cérémonie, H., Su, N., Murphy, C. K. & Youngman, P. (2001). Genome-wide analysis of the general stress response in Bacillus subtilis. Mol Microbiol 41, 757–774.[CrossRef][Medline]

Silva, M. C. & Batt, C. A. (1995). Effect of cellular physiology on PCR amplification efficiency. Mol Ecol 4, 11–16.[Medline]

Sleator, R. D., Gahan, C. G. M., O'Driscoll, B. & Hill, C. (2000). Analysis of the role of betL in contributing to the growth and survival of Listeria monocytogenes LO28. Int J Food Microbiol 60, 261–268.[CrossRef][Medline]

Sleator, R. D., Wouters, J., Gahan, C. G., Abee, T. & Hill, C. (2001). Analysis of the role of OpuC, an osmolyte transport system, in salt tolerance and virulence potential of Listeria monocytogenes. Appl Environ Microbiol 67, 2692–2698.[Abstract/Free Full Text]

Small, P., Blankenhorn, D., Welty, D., Zinser, E. & Slonczewski, J. L. (1994). Acid and base resistance in Escherichia coli and Shigella flexneri: role of rpoS and growth pH. J Bacteriol 176, 1729–1737.[Abstract]

Smoot, L. M., Smoot, J. C., Graham, M. R., Somerville, G. A., Sturdevant, D. E., Migliaccio, C. A., Sylva, G. L. & Musser, J. M. (2001). Global differential gene expression in response to growth temperature alteration in group A Streptococcus. Proc Natl Acad Sci U S A 98, 10416–10421.[Abstract/Free Full Text]

Sue, D., Boor, K. J. & Wiedmann, M. (2003). {sigma}B-dependent expression patterns of compatible solute transporter genes opuCA and lmo1421 and the conjugated bile salt hydrolase gene bsh in Listeria monocytogenes. Microbiology 149, 3247–3256.[CrossRef][Medline]

Takami, H., Takaki, Y. & Uchiyama, I. (2002). Genome sequence of Oceanobacillus iheyensis isolated from the Iheya Ridge and its unexpected adaptive capabilities to extreme environments. Nucleic Acids Res 30, 3927–3935.[Abstract/Free Full Text]

Vandecasteele, S. J., Peetermans, W. E., Merckx, R. & Van Eldere, J. (2001). Quantification of expression of Staphylococcus epidermidis housekeeping genes with Taqman quantitative PCR during in vitro growth and under different conditions. J Bacteriol 183, 7094–7101.[Abstract/Free Full Text]

Venables, W. N. & Ripley, B. D. (2002). Modern Applied Statistics with S. 4th edn. New York: Springer.

Wiedmann, M., Arvik, T. J., Hurley, R. J. & Boor, K. J. (1998). General stress transcription factor {sigma}B and its role in acid tolerance and virulence of Listeria monocytogenes. J Bacteriol 180, 3650–3656.[Abstract/Free Full Text]

Received 20 April 2004; revised 16 July 2004; accepted 13 August 2004.



This Article
Abstract
Full Text (PDF)
Alert me when this article is cited
Alert me if a correction is posted
Citation Map
Services
Email this article to a friend
Similar articles in this journal
Similar articles in PubMed
Alert me to new issues of the journal
Download to citation manager
Google Scholar
Articles by Sue, D.
Articles by Boor, K. J.
Articles citing this Article
PubMed
PubMed Citation
Articles by Sue, D.
Articles by Boor, K. J.
Agricola
Articles by Sue, D.
Articles by Boor, K. J.


HOME HELP FEEDBACK SUBSCRIPTIONS ARCHIVE SEARCH TABLE OF CONTENTS
INT J SYST EVOL MICROBIOL MICROBIOLOGY J GEN VIROL
J MED MICROBIOL ALL SGM JOURNALS
Copyright © 2004 Society for General Microbiology.