Low Concentrations of Bile Salts Induce Stress Responses and Reduce Motility in Bacillus cereus ATCC 14570

Abstract

Tolerance to bile salts was investigated in forty Bacillus cereus strains, including 17 environmental isolates, 11 dairy isolates, 3 isolates from food poisoning outbreaks, and 9 other clinical isolates. Growth of all strains was observed at low bile salt concentrations, but no growth was observed on LB agar plates containing more than 0.005% bile salts. Preincubation of the B. cereus type strain, ATCC 14579, in low levels of bile salts did not increase tolerance levels. B. cereus ATCC 14579 was grown to mid-exponential growth phase and shifted to medium containing bile salts (0.005%). Global expression patterns were determined by hybridization of total cDNA to a 70-mer oligonucleotide microarray. A general stress response and a specific response to bile salts were observed. The general response was similar to that observed in cultures grown in the absence of bile salts but at a higher (twofold) cell density. Up-regulation of several putative multidrug exporters and transcriptional regulators and down-regulation of most motility genes were observed as part of the specific response. Motility experiments in soft agar showed that motility decreased following bile salts exposure, in accordance with the transcriptional data. Genes encoding putative virulence factors were either unaffected or down-regulated.

During passage through the gastrointestinal (GI) tract, ingested bacteria face several challenges such as the acidic environment of the stomach, elevated osmolarity, oxygen starvation, nutrient competition, the immune response, and exposure to a number of different potentially toxic compounds such as bile and degradative enzymes.

Bile is a yellow-green aqueous solution that is produced by the liver and secreted into the upper duodenum (upper small intestine) from the bile duct. The major organic constituents of bile are bile salts and, to a lesser extent, cholesterols, phospholipids, and the pigment biliverdin (18). The concentration of bile salts in the small intestine ranges from approximately 0.2 to 2% (wt/vol), depending upon the individual and the type and amount of food ingested (18). Approximately 95% of the bile salts are reabsorbed by passive diffusion in the length of the small intestine and by active transport in the distal ileum (lower small intestine) (14). In human bile the most abundant bile acids are cholic acid, deoxycholic acid, chenodeoxycholic acid, and lithocholic acid, which constitute approximately 95% of the total (14). The main purpose of bile secretion is to emulsify and dissolve ingested fats, but a significant bactericidal effect is also achieved due to disaggregation of the lipid membrane. Furthermore, bile salts are reported to enter the bacterial cells and cause oxidative stress and damage to DNA (39, 64) and to have induced expression of molecular chaperones in several different bacteria, suggesting that they may also lead to the misfolding of proteins (23, 44, 62).

Studies in both gram-positive and gram-negative bacteria have identified several genomic loci involved in bile (salts) response. The genes involved encode general stress proteins such as molecular chaperons (23, 41, 44, 61), proteins involved in relieving oxidative stress (8, 42), cell wall and membrane proteins (11, 37), and efflux proteins presumably involved in transporting bile salts out of the cell (6, 37, 46, 55). Cross-protection between other types of stress, such as thermal stress or exposure to detergents, and stress from bile exposure has been observed in several bacteria (6, 23). This is in agreement with the induction of general stress proteins in many bile-treated bacteria (22, 41, 62).

The endospore-forming opportunistic pathogenic bacterium Bacillus cereus is a member of the B. cereus group of gram-positive bacteria. Bacteria in this group are genetically closely related and include, among others, both Bacillus anthracis, the causative agent of anthrax in mammals, and Bacillus thuringiensis, an insect pathogen (36). B. cereus is responsible for a wide range of opportunistic infections in humans and is one of the most common causes of bacterial food poisoning. Generally, it causes two types of illness: emesis or diarrhea (reviewed in reference 32). The emetic type is caused by ingestion of the heat-, acid-, and protease-stable peptide toxin cereulide (2, 51) and is characterized by symptoms such as abdominal pain, nausea, and vomiting, usually lasting 1 to 5 h. The diarrheal type, which causes abdominal pain, cramps, and watery diarrhea, is usually experienced 8 to 16 h after ingestion of contaminated food and usually lasts for 12 to 24 h (30). The symptoms are caused by at least one of three enterotoxins: hemolysin BL (HBL), nonhemolytic enterotoxin, and/or cytotoxin K (CytK) (31, 34, 47-49). The B. cereus type strain (ATCC 14579) used in this study is an environmental strain and expresses the genes for the toxins HBL and nonhemolytic enterotoxin and shows promoter activity for cytK under laboratory conditions (10, 29). It is not known whether toxins are produced by ingested vegetative cells or by ingested spores germinating while migrating through the GI tract. Bacteria in the B. cereus group synthesize a range of additional putative virulence factors (phospholipases, collagenases, and proteases) (29), but the molecular mechanism of pathogenesis in these infections is poorly understood with the exception of the action of the emetic toxin and enterotoxins. Although little is know about the ability of B. cereus to colonize the human intestine, this bacterium is frequently isolated from fecal samples (28, 69). The mode of survival is not known, but it has been demonstrated that Bacillus spp. including Bacillus subtilis can germinate and sporulate in the intestinal tract of mice (25, 65).

The molecular responses to heat stress in B. cereus have been investigated by Periago et al. and include induction of typical stress proteins such as GroEL, DnaK, ClpP, and CspB and other proteins (52). As in other gram-positive bacteria these proteins are under the control of the alternative sigma factor, σ^B (52, 70, 71). Some investigations have also evaluated the response of B. cereus to acid, ethanol, and osmolarity, and the induction of several of the same proteins has been observed (13, 52). However, to our knowledge, responses to bile (salts) have not been investigated in B. cereus, and this work describes the first in vitro global transcriptional response to bile salts in this group of bacteria.

METHODS

Bacterial growth, screening for bile-resistant strains, and adaptation.

B. cereus was grown in liquid medium containing tryptone (5 g/liter), yeast extract (2.5 g/liter), glucose (5 g/liter), NaCl (5 g/liter), K₂HPO₄ (0.8 g/liter), KH₂PO₄ (0.2 g/liter), CaSO₄·2H₂O (0.05 g/liter), and MgSO₄·7H₂O (0.25 g/liter) at 37°C with shaking at 175 rpm. Previous work in our laboratory has shown that this modified medium does not give a large increase in the final pH of the growth medium as observed with LB medium. LB-agar plates were used (10 g/liter tryptone, 5 g/liter yeast extract, 10 g/liter NaCl, and 20 g/liter agar) for growth on solid medium.

Forty B. cereus group strains were screened for bile salt tolerance, including 17 environmental isolates, 11 dairy isolates, and 12 clinical isolates, 3 of which have been involved in food poisoning. In addition, six other bacteria not in the B. cereus group were tested, including five nonenteric nonpathogen bacteria and Escherichia coli ATCC 8739. The strains are listed in Table 1. Cells were grown to mid-exponential phase and plated on LB-agar plates supplemented with 0 (control), 0.005, 0.01, 0.05, and 0.2% (wt/vol) bile salts (Na-cholate:Na-deoxycholate, 1:1; Fluka, Buchs, Switzerland). The LB-agar plates were incubated at 37°C, and colonies were counted after 18 and 72 h.

TABLE 1.

Bile salts tolerance of tested bacteria strains

Type of isolate	Bacterial strain(s)a	Growth on LB-agar plates supplemented with bile salts at the indicated concn
		0.005%	0.01%	0.05%	0.2%
Environmental	B. cereus strains ATCC 14579 (type strain), ATCC 6464, ATCC 27348, ATCC 27877, ATCC 33019, ATCC 43881, ATCC 10792, ATCC 53522, ATCC 11778, AH533, AH540, AH557, AH564, AH568, AH648, AH775; B. thuringiensis subsp. kurstaki 4D1	Yes	No	No	No
Dairy	B. cereus strains ATCC 10987, Vet. 31 (AH597), Vet. 4 (AH598), Vet. 5 (AH599), Vet. 8 (AH601), Vet. 9 (AH602), Vet. 10 (AH603), Vet. 11 (AH604), Vet. 1 (AH605), Vet. 59 (AH607), Vet. 61 (AH608)	Yes	No	No	No
Clinical	B. cereus strains AH717, AH722, AH728, AH811, AH812, AH818, AH820, AH830, F4810/72 (emesis, AH187), NVH 391-98 (diarrhea), AH1353 (diarrhea); B. thuringiensis subsp. konkukian 97-27	Yes	No	No	No
Non-B. cereus group bacteria	E. coli ATCC 8739	Yes	Yes	Yes	Yes
	Bacillus sphaericus AH1389, Micrococcus luteus AH1390	Yes	Yes	No	No
	Bacillus megaterium AH1388, Azotobacter vinelandii AH1391, Arthrobacter crystallopoietes AH1393	Yes	No	No	No

^aVet and NVH, strains from the Norwegian School of Veterinary Science. The AH designation is given in parentheses when the AH number has been used in the literature.

The survival of B. cereus ATCC 14579 spores in bile salts was tested by exposure to 0.05, 0.2, or 1% bile salt for 15 min in liquid medium. After exposure the spore suspensions were diluted 10 times in growth medium and plated on LB-agar plates. The plates were examined after a 24-h incubation at 37°C.

Possible adaptation of B. cereus ATCC 14579 to bile salts was evaluated by preexposure of exponentially growing bacteria to bile salts (0.005%) for 5, 30, and 60 min. After preexposure they were challenged with different concentrations of bile salts (0.01, 0.02, 0.05, or 0.1%) for 2 min in liquid medium, after which samples were diluted 10 times in ice-cold growth medium prior to further dilution and plating on LB-agar.

Motility tests.

B. cereus was grown to mid-exponential phase and inoculated, using a thin needle, into LB-soft agar (0.4% [wt/vol] agar) tubes with or without bile salts (0.005%, wt/vol) under both anaerobic (85% N₂, 10% CO₂, 5% H₂) and aerobic conditions and incubated for up to 150 h at 37°C.

Shift experiments.

Growth medium (310 ml) was inoculated with 31 μl of an overnight culture of B. cereus ATCC 14579 and grown in a 3-liter Erlenmeyer flask at 37°C with shaking at 175 rpm to a cell density of ∼5 × 10⁷ cells ml⁻¹ (3 h). Two 100-ml aliquots were transferred to two 1-liter Erlenmeyer flasks containing 100 ml of fresh, preheated medium with or without bile salts to produce a control culture (no bile salts; samples prefixed with C) and bile salts culture (samples prefixed with BS) with bile salts at a final concentration of 0.005% (wt/vol). Samples (10 or 20 ml) for RNA isolation were collected immediately after the shift from the control culture (C0; reference) and after 15, 30, and 60 min from both cultures (from C15, C30, and C60 and from BS15, BS30, and BS60). The time points for sampling are indicated in Fig. 1B.

FIG. 1.

Growth and pH curves of B. cereus ATCC 14579 in the presence or absence of bile salts. After 3 h of growth the culture was shifted to fresh medium with or without bile salts added. Triangles represent the start culture, circles represent the control culture, and squares represent the bile salt culture (0.005%). The diamonds and dotted line represent the cell density in a 0.01% bile salt culture. Solid lines indicate the cell density, and the dashed lines represent the pH. The error bars indicate the standard deviation of at least three separate experiments. (A) Growth from start to stationary phase. (B) Magnification of the time period in which samples were taken for RNA isolation.

The bacteria cell density was measured by flow cytometry using a bacteria counting kit from Molecular Probes (Eugene, OR) and a Becton Dickinson FACSCalibur Flow Cytometer, with the software BD CellQuest Pro (San Jose, CA).

RNA isolation.

Samples were added to an equal volume of ice-cold methanol and incubated at room temperature for 3 to 5 min. Cells were harvested by centrifugation at 4,000 × g at 4°C for 5 min. The cell pellets were stored at −80°C until used. One milliliter of boiling lysis buffer (2% sodium dodecyl sulfate, 20 mM NaCl, 16 mM EDTA, pH 8.0) was added to the frozen cell pellets, and boiling was continued for 3 min before 1 ml of Trizol (Invitrogen Ltd., Paisley, United Kingdom) was added. Further isolation was carried out according to the manufacturer's protocol. The isolated RNA was purified using an RNeasy mini kit from QIAGEN (West Sussex, Great Britain), using the optional on-column DNase treatment step, all according to the manufacturer's instructions. The integrity of the RNA was confirmed by examination on formaldehyde-containing agarose gels (2 μg of RNA loaded). Protein contamination was excluded by measuring the A₂₆₀/A₂₈₀ ratio, and the concentration was determined by measuring the A₂₆₀.

Microarray design.

The cDNAs were hybridized on a microarray array of 70-mer oligonucleotides representing all of the 5,255 (5,234 chromosomal and 21 plasmid-borne) B. cereus ATCC 14579 open reading frames (ORFs) identified by analysis of the genome sequence (GenBank accession numbers AE016877 and AE016878 for chromosome and plasmid, respectively) (38). The 70-mers were designed and synthesized by QIAGEN-Operon (California) and printed on UltraGAPS gamma amino silane-coated slides from Corning (Massachussetts), with the Midigrid Biorobotics arrayer at the Norwegian Radium Hospital (Oslo, Norway). The chip was designed in a composite manner and contained 7,787 features from B. anthracis strains Ames and A2012 (58, 59) and B. cereus strain ATCC 14579 (38). For B. cereus ATCC 14579, 3,645 ORFs that matched a B. anthracis Ames probe with 93% nucleotide identity or more were represented by the B. anthracis Ames oligonucleotide. Probes for the remaining B. cereus ORFs were designed specifically from the B. cereus ATCC 14579 genome sequence. In addition, the arrays had both positive and negative controls and 70-mer oligonucleotides corresponding to Arabidopsis thaliana RNA spikes. The oligonucleotide probes were Tm normalized to a midpoint temperature of 75°C (±5°C). Each 70-mer was replicated in pairs, and each array was replicated once on the slide, giving four replicates of each oligonucleotide on the slide. The microarray is commercially available at http://www.operon.com/arrays/oligosets_anthrax.php.

Microarray sample processing and data collection.

cDNA synthesis and labeling were performed using a FairPlay microarray labeling kit (Stratagene, California). The protocol was followed in all steps except for the use of random hexamer (500 ng; Applied Biosystems, California) primers. Briefly, RNA (20 μg) and spike mixture (Stratagene) were reverse transcribed, purified, and labeled with amino-allyl coupling of Cy3 and Cy5 dyes (Amersham, Uppsala, Sweden). The labeled cDNA was denatured (at 95°C for 2 min) and left at 42°C for up to 20 min before hybridization. Hybridization was done according to the Corning manual (17a). The slides were prehybridized, hybridized overnight at 42°C, and washed before scanning.

The slides were scanned with an Axon 4000B scanner (Molecular Devices Corp., California). Image analysis, gridding, spot annotation, and removal of poor spots were done using the GenePix Pro, version 6.0, software (Molecular Devices Corp.).

cDNA samples prepared from RNA, C15, C30, C60, BS15, BS30, and BS60 were hybridized against a common reference cDNA prepared from RNA from the control isolated immediately after the shift (C0). The reference RNA was pooled from four different experiments. This strategy permitted observation of the transcription trends of each gene for three time points (15, 30, and 60 min) in the bile salt and control cultures. The reason for doing this was that there were significant changes in the growth rates of the two cultures and that after 30 and 60 min of growth the cell density in the control was approximately double that of the bile salt culture. This experimental setup permits discrimination of transcriptional changes caused by the bile salts and changes caused by growth as a function of cell density. All samples were hybridized in three different biological replicates, and each of these was technically repeated by dye swaps. This, added to the fact that each oligonucleotide was repeated four times on each slide, gave 24 measurements for each probe.

Data analysis.

The microarray data were analyzed with a custom R-script in R 2.0.1 (56) (http://www.r-project.org/) with the Limma package (63). The GenePix files were imported to R, and the Limma package was used for filtering, normalization, and further analysis. The mean red and green foreground values and median red and green background values were used. The data were first filtered so that all control spots were weighted to zero, while spots that were flagged as “bad” in GenePix, saturated in both channels, or with weak foreground intensities in both channels were weighted to “no answer.” The saturation limit was set to 60,000. Lower-bound foreground intensity threshold values were set to an intensity of two times background intensity (i.e., a signal-to-noise ratio of 2). Background correction was done using the Edwards method (20). Normalization within each slide was done with the Loess method (16). After normalization, controls were removed, and replicates corresponding to the same gene within each slide were averaged, provided that there were at least two replicates left after the initial filtering procedure. Otherwise, the gene was removed. The average values were compared across slides (biological and dye-swap replicates) by fitting a linear model to the data for each gene (lmFit function). The model was evaluated using empirical Bayesian statistics, and moderate t statistics were computed. On the basis of that, P values were computed using a false discovery rate correction. Genes were considered regulated at a statically significant level if they had a P value below 0.05 and a log₂ ratio above or below ±1 (twofold regulation) relative to the reference cDNA (C0).

Microarray data accession numbers.

An ADF file containing all design and gene information has been deposited in the ArrayExpress database http://www.ebi.ac.uk/arrayexpress under accession number A-MEXP-587. The raw data files are also available in the ArrayExpress database under accession number E-MEXP-896.

RESULTS AND DISCUSSION

Bile salt tolerance and growth.

Forty different B. cereus group strains, including B. cereus ATCC 14579, were screened for bile salt tolerance by plating on LB-agar plates containing different concentrations of bile salts. All of the B. cereus strains tested showed similarly low tolerance to bile salts, i.e., no growth on LB-agar plates supplemented with 0.01% bile salts but growth on plates supplemented with 0.005%. This would appear to be a general property of B. cereus (Table 1). In contrast, E. coli ATCC 8739, isolated from feces, grew on agar plates containing 0.2% bile salts (Table 1). Other enteric pathogens, such as Listeria monocytogenes and Enterococcus faecalis, can survive bile salt concentrations as high as 0.3%, comparable to concentrations found in the small intestine (6, 22). The B. cereus type strain spores were viable after a 15-min exposure to 1% bile salts, and no significant reduction in numbers was observed between treated and untreated spores. The nonenteric, non-B. cereus group bacteria tested also grew on 0.005% bile salt, and two of the bacteria also grew on 0.01% (Table 1). Preexposure of B. cereus ATCC 14579 to bile salts did not give any increase in tolerance levels (data not shown).

B. cereus ATCC 14579 grew at a reduced rate when shifted to medium containing 0.005% bile salts, and growth stopped when the bacteria were shifted to cultures containing 0.01% bile salts (Fig. 1A and B). Control cultures and bile salt cultures (0.005%) reached the stationary phase at the same time but at different final cell densities, i.e., ∼10⁹ cells ml⁻¹ for the control culture and a density five times lower at ∼2 × 10⁸ cells ml⁻¹ for bile salt cultures. Growth curves, in the presence and absence of 0.005% bile salts, were determined for an additional four B. cereus strains. No major differences in the growth patterns were observed (data not shown). The transition to the stationary phase occurred after the time window for RNA sampling, between 60 and 120 min after the shift.

Transcriptional response to bile salts.

Genes showing at least twofold differential expression in one of the bile salt time points compared with the reference (C0, the control culture immediately after the shift from the start culture) and with a confidence level (P value) below 0.05 were selected. Genes showing similar expression in the two 15-min samples (BS15 and C15) were subtracted out. Genes considered to be of particular relevance to the study, based on their annotated function, were also included. The addition of bile salts to culture medium induced 100 genes and repressed 133 genes in samples taken after a 15-min bile salt treatment compared with the reference C0 sample after genes with similar expression in the C15 sample were excluded (Fig. 2). A selection of these genes is listed in Table 2 (for a complete list, see Table S1 in the supplemental material). Generally, two types of responses were seen. (i) Genes associated with bile salt or similar stress types, including several efflux proteins and transcriptional regulators, e.g., drug resistance transporters of the EmrB/QacA family and marR transcriptional regulator (Table 2 and Fig. 3A), were up-regulated. (ii) Typical general stress proteins were up-regulated, and many of these genes were also up-regulated in control cultures but only at higher cell densities, e.g., groES, clpP, and terD (Table 2 and Fig. 3B). Based on the Clusters of Orthologous Groups of proteins (COG) classification (67), classes containing most of the repressed genes were cell motility; cell wall and membrane biogenesis; and DNA replication, recombination and repair (Fig. 2). Classes containing most of the up-regulated genes were carbohydrate transport and metabolism; transcription; and posttranslational modification, protein turnover, and chaperones (Fig. 2). The response to bile salts was most pronounced at 15 min following the start of bile salt exposure, and the transcription levels for many of the differentially regulated genes seemed to even out in both cultures after 60 min (Fig. 3A to D). As much as 38% of the up-regulated and 25% of the down-regulated genes were assigned with general or unknown functions or not assigned in COG.

FIG. 2.

Number of regulated genes, 15 min after the shift to bile salt medium, belonging to each COG class. The red bars represent the number of up-regulated genes, and the green bars represent the number of down-regulated genes.

TABLE 2.

Expression patterns of genes responding to bile salt treatment

COG class and locus tag	Putative function(s) (gene name)	Log2 ratio of culture at indicated time and cell density (no. of cells/ml)a
		Bile salt culture			Control culture
		15 min	30 min	60 min	15 min	30 min	60 min
		4.0e7 ± 1.9e6	5.2e7 ± 9.9e6	1.2e8 ± 2.4e7	5.3e7 ± 9.3e6	1.0e8 ± 2.8e7	2.8e8 ± 7.7e7
Motility
BC1625	Chemotaxis protein (motA)	2.1	0.0	−0.1	0.5	0.4	0.8
BC1627	Chemotaxis protein (cheY), chemotaxis response regulatorb	0.9	0.7	−0.1	−0.1	−0.2	−0.1
BC1639	Flagellar protein (fliS)	−0.8	−0.6	−1.2	0.2	−0.3	0.7
BC1644	Flagellar M-ring protein (fliF)	−2.0	−0.8	−0.1	−0.4	−0.3	0.3
BC1645	Flagellar motor switch protein (fliG)	−1.7	−1.0	−1.0	−0.5	−0.5	−0.3
BC1647	Flagellum-specific ATP synthase	−1.5	−0.9	−0.9	−0.5	−0.5	−0.7
BC1651	Flagellar hook protein (flgE)	−2.5	−0.5	−0.8	−0.3	0.1	1.1
BC1657	Flagellin	−1.4	−0.1	0.1	0.1	0.5	1.4
BC1658	Flagellin	−0.9	0.2	1.4	0.0	0.4	1.3
BC1659	Flagellin	−0.9	−0.1	1.1	−0.2	0.2	1.2
Transport
BC0515	ABC transporter permease protein	1.0	0.1	0.7	−0.5	−0.1	−0.8
BC0684	Di- or tripeptide transporter	1.7	1.0*	1.6	NA	NA	0.0
BC0962	Drug resistance transporter, EmrB/QacA family	1.9	0.7	2.3	NA	NA	−0.2
BC1184	Na+-driven multidrug efflux pump	1.9	2.3	0.8	−0.6	−0.9	−1.1
BC3663	Drug/metabolite exporter family	2.7	1.9	1.5	0.4	0.7	0.2
BC4000	Drug resistance transporter, EmrB/QacA family	2.7	2.8	2.7	0.1	0.1	−0.7
BC4568	Drug resistance transporter, EmrB/QacA family	4.6	3.8	4.1	NA	NA	1.0
BC4569	Drug resistance protein A	3.1	2.7	2.2	NA	NA	0.5
BC4707	Drug resistance transporter, EmrB/QacA family	1.3	1.0	0.8	−0.2	−0.3	−0.3
BC5399	ABC transporter, permease protein	1.6	0.6	1.2	0.4	0.1	1.2
Regulators
BC1004	Sigma Bb	0.4	0.7	1.2*	−0.3	0.5	1.1
BC0099	Transcriptional regulator (ctsR)b	1.6**	1.0**	1.3**	0.7	0.8	2.2
BC4603	Transcriptional regulator, GntR family	−1.4	−0.3	−0.4	0.3	0.1	0.1
BC1302	Transcriptional regulator, GntR family	1.0	0.8	0.8	0.0	0.1	0.5
BC0801	Transcriptional regulator, LytR family	−1.1	0.0	0.2	0.0	−0.3	−0.2
BC0657	Transcriptional regulator, MarR family	3.0	2.6	3.0	0.1	−0.1	0.8
BC0954	tcdA-E operon negative regulator	1.6	1.0**	0.3	0.4	0.2	0.6
BC3160	Transcriptional regulator, TetR family	1.2	0.8	0.8	NA	0.1	0.2
BC1814	Transcriptional regulator, TetR family	1.4	0.5	0.9	0.3	−0.1	−0.1
BC2936	Transcriptional repressor (bm3R1)	1.5*	1.1	0.4	0.1	0.1	−0.4
BC5350	Transcriptional activator (plcR)b	0.0	0.2	0.7	0.2	0.8	2.3
BC4314	Heat-inducible transcription repressor (hrcA)b	0.6	0.1	0.5	0.3	0.6	1.2
BC3601	Two-component response regulator	−2.2	−0.6	0.5	−0.1	0.1	1.5
Stress
BC0294	Chaperonin (groES)	2.3	2.0	3.1	−0.1	0.8	2.1
BC0295	Chaperonin (groEL)	1.7	1.3	2.7	0.0	0.5	2.0
BC5152	ATP-dependent Clp protease, proteolytic subunit (clpP)	2.7	1.8	1.9	0.0	0.7	1.9
BC1168	ATP-dependent Clp protease, ATP-binding subunit (clpB)	2.0	1.1**	1.3*	0.3	0.6	2.3
BC4859	Cold shock protein (cspD)	2.7	2.5	1.4*	0.6	0.7	0.5
BC2214	Small heat shock protein (hsp20) family	4.1	0.8	1.3	NA	0.4	2.0
BC4272	Superoxide dismutase	1.5	1.3	0.1	−0.1	−0.5	−0.6
BC0443	Tellurium resistance protein (terD)	1.4	0.8	1.5	0.0	0.7	2.1
BC4521	Thioredoxin	0.9	1.6	1.4	−0.6	−0.8	1.3
BC5159	Thioredoxin reductase	1.5	0.7	0.7	−0.2	−0.2	0.8
BC1774	Peptide methionine sulfoxide reductase	2.1*	1.1*	1.6	0.7	0.7	0.9
Virulence
BC3103	Hemolysin BL lytic component L1	−1.5*	−0.4	2.5	−0.2	3.0	4.6
BC1331	Internalin, putative	−2.5	−1.7	−1.8	−0.6	−1.9	−3.1
BC4377	Collagenase	−2.1*	−2.0	−1.7	−0.3	−0.8	−2.5
BC5101	Perfringolysin O precursor thiol-activated cytolysin	−2.0	−0.8	0.9	0.3	2.3**	4.9
BC0671	Sphingomyelinase C	−1.6	0.0	1.4	−0.3	1.6**	4.6
Cell wall
BC1974	Peptidoglycan N-acetylglucosamine deacetylase	−1.5	−0.6	−0.7	0.1	0.1	0.0
BC5271	UDP-N-acetylglucosamine 4-epimerase	−2.8	−1.4	−0.8	−0.4	−0.1	−0.8
BC5273	UDP-bacillosamine synthetase	−2.7	−0.9	−0.3	0.1	−0.6	−1.2
BC5274	UDP-N-acetylglucosamine 4,6-dehydratase	−2.9	−1.7	−1.3	−0.1	−0.7	−1.3
BC5429	Beta-1,3-N-acetylglucosaminyltransferase	−1.5	−0.4	−1.1	−0.2	−0.5	−1.4
BC0740	Cell wall endopeptidase, family M23/M37	−1.8	−0.7	0.3	0.1	1.0	0.5
BC0991	S-layer protein, putative	−1.5	0.0	0.3	0.1	1.1**	0.8
Fatty acid, biosynthesis and degradation
BC0400	Fatty acid desaturase	1.2	1.0	0.3	0.5	0.1	−1.7
BC1788	Lysophospholipase L2	1.2	0.2	0.3	0.3	−0.1	0.9
BC3781	CDP-diacylglycerol-glycerol-3-phosphate 3-phosphatidyltransferase	1.2	−0.3	NA	NA	−0.4	0.4
BC4602	Acetyl-coenzyme A carboxylase, subunit beta	−1.0	−0.6	−0.7	0.4	0.2	−0.1
Metabolism; glycolysis, fermentation, and electron transport
BC5320	PTS system, glucose-specific IIA component	1.5	1.0	0.7	−0.1	0.4	−0.3
BC0668	(R,R)-butanediol dehydrogenase	2.2	1.9	2.5	0.2	1.4*	4.1
BC4599	Pyruvate kinase	−2.2	−1.1*	0.1	−0.3	−0.2	0.8
BC4600	6-phosphofructokinase	−1.4	−1.2*	−0.4	−0.4	−0.2	0.6
BC4962	Fructose-1,6-bisphosphataseb	0.9	0.0	−0.5	−0.2	−0.4	−0.1
BC3721	NADH-dependent butanol dehydrogenase A	2.0	0.4	1.6	0.3	0.4	1.5
BC1376	Flavodoxin	−2.1	−2.3	−2.7	−0.7	−2.0**	−3.9
Protein synthesis
BC0135	Small-subunit ribosomal protein S19P	−2.0	−1.1**	−1.7*	0.0	−0.2	−1.3
BC5075	Large-subunit ribosomal protein L11P	1.4	0.9	1.1	0.2	0.6	0.5
BC0154	Translation initiation factor 1 IF-1	−2.0	−1.0*	−0.4	0.1	−0.1	−0.8
BC3864	Methionyl-tRNA formyltransferase	−2.7	−0.7	−0.6	0.3	−0.2	−0.4
BC4561	Phenylalanyl-tRNA synthetase alpha chain	−3.3	−1.7	−0.3	−0.2	0.3	0.0
BC3714	tRNA delta(2)-isopentenylpyrophosphate transferase	1.7	0.5	−0.1	0.0	0.4	0.4
DNA replication, repair, recombination, nucleotide synthesis
BC1921	Site-specific recombinase	−1.5	0.2	0.9	0.2	0.0	1.3
BC3769	DNA mismatch repair protein mutS	−2.2	−1.2	0.0	−0.1	0.0	0.3
BC4487	Superfamily I DNA and RNA helicases	−1.8	−0.6	0.3	−0.2	−0.2	0.2
BC0941	Type II restriction-modification system methylation subunit	−1.8	−1.0	−1.0	0.4	0.7	1.3
BC4456	Type I restriction-modification system restriction subunit	−3.1	−0.7	1.0	−0.2	0.3	0.1
BC4459	Type I restriction-modification system methylation subunit	−2.1	−0.2	1.2	0.2	0.3	−0.1
BC5272	Carbamoyl-phosphate synthase small chain	−2.2	−1.0	−0.8	0.0	−0.3	−0.3
BC3885	Dihydroorotate dehydrogenase electron transfer subunit	−2.8	−1.2*	−1.5	−0.8	−0.1	−2.7
BC3887	Carbamoyl-phosphate synthase, small chain	−2.4	0.1	−1.2	−0.7	0.7	−2.0
Other
BC5087	Putative lantibiotic precursor peptide	2.4	2.2	3.8	0.3	1.2*	2.8
BC5088	Putative lantibiotic precursor peptide	2.3	2.7	4.1	0.1	1.5	2.8

^aTime is for the period after the cultures were shifted to medium with or without bile salts; cell density is the value at the time of harvest. The log₂ ratio for the bile salt cultures is calculated between each sample and the reference RNA (cDNA). NA, no answer (no signal above two times the background level was observed in either channel or the gene was filtered out). *, P value between 0.05 and 0.10; **, P value above 0.10.

^bThese genes are not significantly regulated in bile salt cultures, but are included because they are relevant to the study.

FIG. 3.

Expression profiles of selected genes relative to the cell density at the time of harvest. The y axis displays the regulation given as the log₂ ratio, and the x axis displays the cell density of the culture at the time of harvest. The expression of each gene is given as the log₂ ratio compared to the reference RNA (C0), which is set to 0 by default. The solid lines and filled symbols represent the bile salt cultures, and the dashed lines and open symbols represent the control culture. (A) Expression trends of mRNA for drug resistance transporters and marR family regulator relative to the cell density at the time of harvesting. •, BC4000 emrB/qacA family drug resistance transporter; ▪, BC4568 emrB/qacA family drug resistance transporter; and ⧫, BC0657, marR family regulator. No signals were obtained for BC4568 in control cultures at 15 and 30 min, and this line is therefore represented by dots. (B) Expression trends of stress genes relative to the cell density at the time of harvesting. •, BC4272 superoxide dismutase; ▪, BC4859 cspD; ▴, BC0294 groES; and ⧫, BC5251 clpP. (C) Expression trends of virulence genes relative to the cell density at the time of harvest. •, BC3103 HBL; ▪, BC5101 perfingolysin O; ▴, BC1331 internalin; and ⧫, BC5350 plcR. (D) Expression trends of genes involved in metabolism of sugars relative to the cell density at the time of harvest. •, BC0668 butanediol dhg; ▪, BC4599 pyruvate kinase; ⧫, BC1376 flavodoxin; ▴, BC5320 PTS, glucose specific.

Motility.

Many of the motility-associated genes in the B. cereus ATCC 14579 genome are located in a ∼45-kb region ranging from locus BC1625 to locus BC1671 (38). Most of these gene transcripts were down-regulated after 15 min in the bile salt cultures relative to the control culture at the same time point (15 min after shift). The expression pattern for the region, shown in Fig. 4, shows that the trend of down-regulation is clear, although not all the genes in the region meet the twofold regulation cutoff. Both motA and cheY display a slight up-regulation. CheY is a chemotactic response regulator responsible for the switch between tumbling and smooth swimming and may be important in moving the cell away from the bile salts, even though the de novo synthesis of the chemotactic machinery is reduced. In the control culture most of the motility genes are not affected or up-regulated relative to the control (C0), suggesting that the reduction of transcript levels for motility genes is specific to bile salts. The down-regulation of flagella genes in bile salt cultures may serve two purposes. The cells are responding to stress and need to concentrate transcriptional and translational capacities on other functions. Alternatively, from an infection point of view, it is possible that the down-regulation of surface molecules is a host evasion strategy because such genes are often encountered as antigens by the host's immune system. This also correlates with the down-regulation of genes encoding other surface proteins such as internalin and collagen adhesion proteins (Table 2).

FIG. 4.

Regulation of a 45-kb region of the B. cereus ATCC 14579 genome containing motility-associated genes in bile salt culture (red and green bars; BS15) and control culture (open bars; C15) 15 min after the shift compared to the reference isolated immediately after the shift (C0). The bars represent the log₂ up-regulation (red) or down-regulation (green) of the displayed genes. Some of the genes listed may have a P value higher than 0.05. These are included because they belong to an operon which has significantly regulated genes or because they are only slightly regulated. UDP-N-Ace, UDP-N-acetylenolpyruvoylglucosamine reductase; Hook 1, -2, -3, flagellar hook-associated protein 1, -2, and -3; ATP synth, flagellum-specific ATP synthase; −, located on the negative strand. Other genes are designated by their Refseq gene names (NC_004722).

Light microscopy observations, during experiments, showed that the motility in the bile salt cultures was reduced. To confirm this, the motility in soft agar tubes, with or without bile salts added, was determined. The reduced motility observed with the microscope and the reduced transcript levels for motility-related genes were, indeed, supported by a reduced motility observed by the soft agar motility test (Fig. 5A). The reduction in motility seemed to be greater in the lower parts of the tube, where the oxygen concentration was assumed to be lower. The experiment was repeated under anaerobic conditions, and it was observed that motility was completely halted whereas motility was retained in the absence of bile salts (Fig. 5B). An expression study of Salmonella enterica serovar Typhimurium by Prouty et al. also reports down-regulation of genes encoding flagellar components by bile stimulation (55). However, as also discussed below, different bacteria respond differently to bile, and the converse is seen in Vibrio cholerae, where transcription levels of motility genes under bile salt stimulation are increased (37).

FIG. 5.

(A) Aerobic motility test. B. cereus ATCC 14579 was incubated aerobically for 24 h in LB-soft agar with or without bile salts added. Tubes show no inoculate (left), culture without bile salts (middle), and culture with bile salts (right). No changes in motility were observed after 24 h. (B) Anaerobic motility test. B. cereus ATCC 14579 was incubated anaerobically for 24 h in LB-soft agar with or without bile salts added. Tubes show no inoculate (left), culture without bile salts (middle), and culture with bile salts (right). No changes in motility were observed after 24 h.

Transcription trends of selected gene classes relative to cell density.

The expression trends for selected genes in some of the most significantly regulated classes are shown in Fig. 3A to D. Genes for which the products are involved in transport, regulation, virulence, general stress response, and metabolism were included. In these figures the cell density was plotted on the x axis, rather than time, such that comparisons could be made on the basis of cell density at the time of harvesting.

Efflux proteins.

Several genes encoding efflux proteins showed increased transcription levels (Fig. 3A and Table 2). It is plausible that several of these may be involved in bile salt efflux. The genes encoding drug resistance transporters of the EmrB/QacA family are induced in bile salts but not in the control, suggesting that this effect is specific to bile salt stimulation. This class of transporters has seven genes in the B. cereus ATCC 14579 genome, and four of these are up-regulated from 2.5-fold to 25-fold after a 15-min bile salt treatment (Table 2), suggesting an important role for these transporters in the bile salt response. The EmrB family of drug resistance proteins has previously been reported to increase resistance to deoxycholate in V. cholerae (17). Despite this, the transporters induced in B. cereus display only weak amino acid similarity with other transporters responding to bile in other bacteria (6, 11, 46, 61). However, the amino acid similarity for the transporters in full is often not as important as the active site for bile salt binding. The QacA family is often regulated by TetR family regulators (12), which are also observed to be up-regulated in the bile salt cultures (Table 2). Interestingly, another gene encoding a transcriptional regulator classified in the MarR family is regulated in much the same pattern as these transporters (Fig. 3A). This multiantibiotic resistance regulator is often involved in resistance toward antibiotics and other organic compounds (3, 4), and it may be that this regulator controls part of the bile salt response seen in B. cereus ATCC 14579. It should also be mentioned that a putative bile salt transporter (BC3503) is encoded in the B. cereus ATCC 14579 genome. However, this transporter is truncated, and no expression changes were observed.

General stress response.

Transcripts for proteins involved in relieving stress caused by misfolded proteins, such as clpP and groESL, were induced by bile salts (Fig. 3B and Table 2). Several of these stress proteins are also increased in other bile (salt)-treated bacteria (22, 26, 44, 60, 62). This correlates well with the assumption made by Begley et al. that bile salts, on entering cells, cause misfolding of proteins due to the detergent properties of bile salts (7). It could also be mentioned that the Clp proteins seems to play a major role in virulence in several other gram-positive bacteria (27, 50, 53). In addition, genes involved in protection against oxidative stress, such as superoxide dismutase and thioredoxins, were up-regulated. Genes for such proteins have been documented to be important for the bile (salt) response in E. coli and Propionibacterium freudenreichii (8, 44), and it has also been shown that bile acids inside cells generate oxidative stress by generating free radicals (64). Interestingly, several general stress response genes (groESL, clp, tellurium resistance proteins, thioredoxin, and hsp20) were also induced in the control cultures relative to C0 but only in the growth phase proximal to the stationery phase, where a reduction in pH occurred (Fig. 1A and B and Table 2). This high level of up-regulation of genes encoding several stress proteins in the control culture could be a response to the reduction in pH starting 30 min after the shift but could also be a response to several other factors, such as high cell density, accumulation of waste products, or nutrient limitation. Other genes encoding stress response proteins such as superoxide dismutase and cspD were not induced in the control cultures at higher cell densities, suggesting that these genes do not take part of the more general stress response seen in the control cultures after 60 min and are probably specific for other types of stresses, including bile salt (Fig. 3B and Table 2). The response to heat, previously reported for B. cereus, has elements in common with the bile salt response; proteomic data confirms the up-regulation of GroEL, Hsp20, superoxide dismutase, ClpP, and thioredoxin in response to heat (52). Several of these proteins were also induced by stimuli such as ethanol, increased osmolarity, low pH, and low temperature (52), underlining the importance of these proteins in multiple types of stresses. This study now confirms their importance in bile salt stress as well. Several of these genes are under the regulation of the principle stress response sigma factor in gram-positive bacteria, σ^B (52, 70, 71). The lack of significant changes in the σ^B mRNA levels could imply that these genes are under the control of other stress response regulators like the CtsR and/or HrcA transcriptional regulators, which are known to regulate stress response in other bacteria (see reference 35 for a review). The expression of these regulators is up-regulated in the control cultures after 60 min (Table 2). The expression of ctsR could also be increased the bile salt cultures, but the P values are too high to draw any significant conclusions (Table 2).

The induction of tellurium resistance genes seen both as a response to bile salt and in the control culture at higher cell densities supports previous suggestions that these genes confer resistance toward compounds other than tellurium and are part of a more general stress response (see references 40 and 68 for a review). Indeed, the expression pattern for these genes resembles the expression pattern for several other general stress genes (Table 2), and it may be that these genes are under a common control element.

Virulence.

In some intestinal pathogenic bacteria, such as V. cholerae, E. coli, and Shigella spp., bile appears to be a localization signal to the intestine, inducing preparations for invasion or virulence (19, 37, 54). This does not, however, apply to all intestinal bacteria; S. enterica serovar Typhimurium represses invasion genes and motility genes upon bile stimulus (55). No increases (above the levels observed in control cultures) in transcription were observed for known virulence genes in B. cereus in the bile salt cultures (Fig. 3C). In general, the expression of virulence genes, such as hbl and perfingolysin O, is reduced following a 15-min bile salt exposure, while after 30 min the levels are the same as those observed in the control culture (Fig. 3C and Table 2). It would appear that B. cereus may be using a more Salmonella-like approach to bile, at least with regard to repression of virulence genes and motility genes.

In the B. cereus group of bacteria, extracellular virulence factors are regulated by a quorum sensing-like system which includes the pleiotropic transcriptional regulator PlcR (1, 29). The mRNA expression levels for this gene do not decrease after 15 min, as observed for the virulence genes, and the transcription of this regulator increases as the growth phase approaches the stationary phase, between 30 and 60 min after the shift in bile salts, as well as in control cultures (Fig. 3C). This correlates with other studies showing that transcription of plcR is induced at the onset of the stationary phase (43). The differences in transcription trends for plcR-regulated genes (29) and expression of plcR itself indicate that these virulence genes are not solely under PlcR control but can also be regulated by other factors.

Metabolism.

There are several changes in the transcription of metabolic genes upon bile salt treatment in B. cereus ATCC 14579. Transcription levels of the genes encoding enzymes in the butanediol pathway are up-regulated, while the pyuvate dehydrogenase complex, coupling glycolysis to the tricarboxylic acid cycle, is down-regulated (Fig. 3D and Table 2). Another gene encoding a butaneol dehydrogenase is also up-regulated and may involve the conversion of acidic fermentation products to neutral products. The reason for using butanediol as a fermentation product is not known, but it has been suggested to play a role in preventing intracellular acidification by changing the metabolism from acid production to the formation of a neutral compound (9). This pathway is also highly up-regulated in the control culture after 60 min, so such an explanation seems reasonable, taking into account the lower pH observed at this time point (Fig. 1A and B). A change toward fermentation is also seen in bile salt-treated Bifidobacterium longum, although in this study the switch was toward other fermentation products (61). It is also interesting to observe an increase in the transcripts for a phosphate transferase system (PTS) importing glucose, together with the up-regulation of genes encoding steps in the gluconeogenesis pathway (fructose-1,6-bisphosphatase) and down-regulation of genes encoding glycolytic enzymes (pyruvate kinase and 6-phosphofructokinase), suggesting alternative routes for glucose utilization (Table 2). The same study on B. longum also reports an increase in glucose consumption (61), which fits well with our data showing the up-regulation of this PTS. From the transcriptional data it appears that energy metabolism in B. cereus ATCC 14579 moves to fermentation rather than respiration upon bile salt stimulation.

GntR-family regulators are known to be involved in a variety of different cellular processes, e.g., in regulation of gluconate metabolism in B. subtilis (24) and fatty acid metabolism in E. coli (74), and are important for virulence in Brucella melitensis (33). In addition, a gntR regulator is deleted in a bile salt-sensitive E. faecalis mutant (41). Bile salt stimulation leads to both up- and down-regulation of gntR regulators in B. cereus ATCC 14579 (Table 2). The gntR regulator (BC4603) and the acetyl-coenzyme A carboxylase (BC4602) are repressed by bile salts and are predicted to be encoded in the same operon (data not shown; www.microbesonline.org), and it is possible that this regulator plays a role in fatty acid metabolism in B. cereus ATCC 14579. Bile salts have a destructive effect on cell membranes, and it is possible that the regulation of genes involved in lipid metabolism (Fig. 2 and Table 2) is a response to this effect. It has previously been shown that adjustment of the fatty acid composition of the membrane concurs with higher bile salt tolerance in Lactobacillus spp. (21, 66).

Concluding remarks.

All of the B. cereus strains tested showed a low tolerance to bile salts (no growth observed on agar plates with 0.01% bile salts). Several intestinal bacteria tolerate high levels of bile (salts), and several proteins involved in DNA repair and cell wall synthesis have been implicated in this response (6, 41, 44, 45). These gene classes are mostly down-regulated in B. cereus ATCC 14579 (Fig. 2 and Table 2), and this may offer a partial explanation for the low tolerance to bile salts observed in B. cereus strains. Two of the B. cereus strains investigated are associated with diarrheal food poisoning. There may be several possible explanations for this phenomenon; B. cereus food poisoning may result from the ingestion of spores which germinate in the more distal parts of the small intestine, where the concentration of bile salts is lower. There are several studies pointing out this possibility; a study by Clavel et al. shows that vegetative B. cereus cells at best survive a pH of 3.6 when incubated in milk (15), while spores tolerate a pH lower than 2. Further, it has been shown that vegetative cells will not survive through the GI tract of human-flora-associated rats, while the spores do survive in the intestine of these rats (73). Finally, experiments recently published show that spores germinate in medium simulating the intestinal environment and that mesophilic strains germinate better than psychrotrophic strains in such medium (72). In this study we have also shown that the spores survive a high concentration (at least 1%) of bile salts. It is not known, however, how a high bacterial load in a potentially protective food matrix may influence survival of vegetative cells in vivo. Furthermore, previous studies verify that both spores and vegetative cells can attach to epithelial cells (5, 57), leaving open the possibility that both vegetative cells and/or spores may be the infectious form(s).