Search tips
Search criteria 


Logo of aemPermissionsJournals.ASM.orgJournalAEM ArticleJournal InfoAuthorsReviewers
Appl Environ Microbiol. 2010 February; 76(4): 1232–1240.
Published online 2009 December 28. doi:  10.1128/AEM.02317-09
PMCID: PMC2820966

Identification of the Main Promoter Directing Cereulide Biosynthesis in Emetic Bacillus cereus and Its Application for Real-Time Monitoring of ces Gene Expression in Foods[down-pointing small open triangle]


Cereulide, the emetic Bacillus cereus toxin, is synthesized by cereulide synthetase via a nonribosomal peptide synthetase (NRPS) mechanism. Previous studies focused on the identification, structural organization, and biochemical characterization of the ces gene locus encoding cereulide synthetase; however, detailed information about the transcriptional organization of the ces genes was lacking. The present study shows that the cesPTABCD genes are transcribed as a 23-kb polycistronic transcript, while cesH, encoding a putative hydrolase, is transcribed from its own promoter. Transcription initiation was mapped by primer extension and rapid amplification of cDNA ends (RACE). Deletion analysis of promoter elements revealed a main promoter located upstream of the cesP coding sequence, encoding a 4′-phosphopantetheinyl transferase. This promoter drives transcription of cesPTABCD. In addition, intracistronic promoter regions in proximity to the translational start sites of cesB and cesT were identified but were only weakly active under the chosen assay conditions. The identified main promoter was amplified from the emetic reference strain B. cereus F4810/72 and fused to luciferase genes in order to study promoter activity in complex environments and to establish a biomonitoring system to assess cereulide production in different types of foods. ces promoter activity was strongly influenced by the food matrix and varied by 5 orders of magnitude. The amount of cereulide toxin extracted from spiked foods correlated well with the bioluminescence data, thus illustrating the potential of the established reporter system for monitoring of ces gene expression in complex matrices.

Bacillus cereus is the causative agent of two types of food poisoning: diarrhea and emesis. The toxicoinfection referred to as diarrheal disease occurs after consumption of B. cereus spores or vegetative cells, most likely due to the action of heat-labile enterotoxins, which are produced by cells multiplying in the small intestine (4, 22, 30, 39). Contrarily, the emetic type of food-borne illness is caused by intoxication with the heat-stable peptide cereulide, which is preformed in foods and elicits vomiting a few hours after ingestion (18, 51, 58).

Regulation of enterotoxin expression of B. cereus has been studied in some detail (e.g., references 16, 29, and 46; for a review, see reference 58), revealing a major role of the pleiotropic transcription regulator PlcR in activation of enterotoxin genes and other virulence factors (1, 29, 45). Cereulide synthesis in emetic B. cereus, in contrast to synthesis of B. cereus enterotoxins, is not controlled by PlcR but by the Spo0A phosphorelay. The global transition state factor AbrB was identified as one factor repressing cereulide production in early exponential phase (38). Although B. cereus enterotoxins and the emetic toxin cereulide seem to belong to completely different regulatory networks, production of both types of toxins depends substantially on nutritional and environmental factors (3, 46, 55), and it is expected that great effort is required to unravel in detail the intrinsic and extrinsic factors and mechanisms controlling toxin expression in B. cereus.

Cereulide, which is produced by specific subgroups of B. cereus (20, 62) and a few Bacillus weihenstephanensis isolates (60), is a small, acid- and proteolytically stable cyclic dodecadepsipeptide that is structurally related to the potassium ionophore valinomycin (2). It is toxic to mitochondria and has been implicated in severe forms of food poisoning resulting in acute liver failures (14, 41, 48). The peptide toxin [d-O-Leu-d-Ala-l-O-Val-l-Val]3 is synthesized enzymatically by a nonribosomal peptide synthetase (NRPS) (21).

NRPSs are responsible for catalyzing the synthesis of a broad range of bioactive low-molecular-weight peptides, with chain lengths of 2 to 48 residues (42). Several Bacillus species produce siderophores and peptide antibiotics via NRPSs, such as the broadly distributed bacillibactin-related siderophores or surfactins or the species-specific antibiotics tyrocidine and gramicidin S in Bacillus brevis (for a review, see references 13, 24, and 57); however, few transcriptional analyses of the corresponding biosynthetic enzyme complexes have been performed.

The complete genetic locus encoding cereulide biosynthetase (ces) in the emetic B. cereus reference strain F4810/72 is a 24-kb gene cluster located on a megaplasmid (pBCE) that shows high homology to the B. anthracis toxin plasmid pXO1 (17). Besides the typical genes, such as a gene encoding a 4′-phosphopantetheinyl (4′-PP) transferase (cesP) essential for priming the NRPS, a gene encoding a putative type II thioesterase (cesT) which removes misprimed monomers, and the structural genes responsible for the assembly of the peptide product (cesA and cesB), it includes a coding sequence (CDS) encoding a putative hydrolase (cesH) in the 5′ region and a putative ABC transporter (cesC and cesD) in the downstream part (17). The cereulide NRPS is unique in that the substrates for the cesA1 (d-O-Leu) and cesB1 (l-O-Val) modules are α-keto acids which are then chirally reduced (40).

Although the genetic locus responsible for cereulide production has been identified and partially characterized, information about the transcriptional organization of the ces operon is still lacking. Such information is crucial for the understanding of cereulide toxin biosynthesis and might also contribute to the fundamental knowledge of nonribosomal biosynthetic pathways. A complete transcriptional analysis of the ces gene cluster was carried out by using reverse transcription-PCR (RT-PCR), rapid amplification of cDNA ends (RACE), and promoter deletion analysis to identify the promoters and characterize the ces transcripts. Our studies revealed a central promoter that drives the polycistronic transcript of cesPTABCD. This promoter sequence was used to establish a bioluminescence reporter system for noninvasive real-time monitoring of cereulide synthetase promoter activity in different environments. This is a first step toward deciphering conditions at a transcriptional level that are favorable or unfavorable for emetic toxin production in food. Such data may contribute to reducing the risk and incidence of food-borne disease.


Bacterial strains and culture conditions.

The emetic reference strain Bacillus cereus F4810/72 (AH187) (61) was routinely cultivated on standard plate count agar (5 g peptone, 2.5 g yeast extract, 1 g glucose per liter) at 30°C. For experiments with food, B. cereus F4810/72 transformed with pMDX[P1/luxABCDE] was cultured in lysogeny broth (LB) (5) for 16 h at 30°C with 150-rpm rotary shaking. Escherichia coli was routinely grown at 37°C in LB. When appropriate, chloramphenicol (5 μg ml−1) or ampicillin (100 μg ml−1) was added.

Sequence analysis.

Putative transcriptional start sites and transcription factor binding sites were searched using promoter prediction at (53) and the DBTBS search tool at (32). Termination structures were analyzed using the Heidelberg Unix Sequence Analysis Resource (7) at and Mfold (64) at Frameshift slippery sites were analyzed using FSFinder at (44).


PCR was used to amplify putative promoter regions both for cloning into vectors to produce sequencing ladders for primer extension and for cloning into fusion vectors to test promoter activity. The 50-μl PCR mixture (10 ng DNA, 0.5 μM each primer, 1.5 mM MgCl2, 0.4 mM each deoxynucleoside triphosphate [dNTP], 1.25 U ThermoStart Taq polymerase [all from ABgene]) was activated (95°C for 15 min), followed by 30 amplification cycles (95°C for 30 s, 60°C for 45 s, and 72°C for 1 min) and an elongation step (72°C for 5 min).

Nucleic acid isolation.

Total DNA was isolated as described previously (19), and plasmid DNA was prepared using standard procedures. RNA samples for RT-PCR, primer extension, and 5′ RACE were taken during mid-exponential-phase growth: cells were collected (10,000 × g at 4°C for 2 min), frozen in liquid nitrogen, and stored at −80°C. Total RNA was isolated by an adapted plant tissue protocol (37) after cell disruption by using 0.1-mm zirconia-silica beads (Carl Roth GmbH & Co. KG) and a RiboLyser (Hybaid). Contaminating DNA was removed with 10 U RQ1 DNase (Promega). Subsequently, RNA was isolated by chloroform extraction and ethanol precipitation, resuspended in diethyl pyrocarbonate (DEPC)-treated double-distilled water (ddH2O), and stored at −80°C. RNA purity and quantity were determined by measurement of absorbance at 260 nm and 280 nm and by agarose gel electrophoresis. Efficacy of DNA removal was confirmed using RNA as the PCR template with 16S primers 16SA1 (5′-GGAGGAAGGTGGGGATGACG-3′) and 16SA2 (5′-ATGGTGTGACGGGCGGTGTG-3′) (43).

RT and RT-PCR.

RT-PCR was used to investigate the ces transcript. cDNA was synthesized using SuperScript III reverse transcriptase (Invitrogen). In 10 μl, 100 ng total RNA, 2 pmol gene-specific reverse primer (see Table S1 in the supplemental material), and 1 μl 10 mM each dNTP were incubated at 75°C for 2 min, 70°C for 1 min, 65°C for 1 min, 60°C for 1 min, and 55°C for 1 min. At 55°C, 10 μl of RT mix (4 μl 5× RT buffer, 1 μl 0.1 M dithiothreitol [DTT], 100 U SuperScript III, 20 U RNaseOUT [all from Invitrogen]) was added, and the reaction mix was incubated at 55°C for 1 h, before inactivation at 70°C for 15 min. For negative controls, reverse transcriptase was omitted. Subsequent PCRs using a 2-μl RT reaction volume (template) with 50 pmol each primer (Table S1) in 50 μl were carried out as described above. Fragments of expected lengths greater than 2 kb were amplified using the Expand high-fidelity PCR system (Roche Applied Science) by following the manufacturer's instructions. All RT-PCR experiments were repeated with RNA samples from several independent cultures.

Transcript stability.

To investigate the stability of the cesA transcript, cultures were grown to mid-exponential phase and samples were taken at time zero (before rifampin addition) and in 10-min increments (up to 60 min) after the addition of 200 μg ml−1 rifampin. RNA was isolated, and cDNA synthesized as described above was used as the template for real-time PCR. Real-time quantitative PCR (qPCR) was performed using a SmartCycler (Cepheid). Reaction volumes contained 1 μl cDNA (equivalent to 10 ng total RNA) and 80 nM each primer in qPCR master mix with SYBR green I (ABgene) in a total volume of 25 μl. Activation (15 min at 95°C) was followed by 40 amplification cycles with a temperature ramp rate of 2.5°C s−1 (95°C for 30 s; melting temperature [Tm] for 30 s with optics on; 72°C for 45 s) and a melt curve analysis from the Tm to 95°C at 0.2°C s−1 with optics on. (The Tm for cesA is 53°C, and that for the 16S rRNA gene is 63°C.) Relative expression was determined using primers for ces (cesA_for [5′-GATTACGTTCGATTATTTGAAG-3′] and cesA_rev [5′-CGTAGTGGCAATTTCGCAT-3′]) and normalized to 16S rRNA genes by using primers previously reported (43). Relative expression was calculated by the REST (relative expression software tool) method of analysis (47), using time point zero as the calibrator.


RACE was performed to find transcriptional start sites. The 5′ RACE System for Rapid Amplification of cDNA Ends, version 2.0 (Invitrogen), was used with the indicated primers (see Table S2 in the supplemental material); cDNA was created using SuperScript III as described above. RACE products were visualized by agarose gel electrophoresis, TOPO TA cloned (Invitrogen), and sequenced to determine the transcription start site. All RACE experiments were repeated several times with RNA samples from independent cultures.


Transcriptional start sites were confirmed using primer extension (PE). cDNA was synthesized by using 10 pmol IRD800-labeled gene-specific reverse primer (Table S2) and 10 μg RNA in a 7-μl reaction volume and incubating at 75°C for 2 min, 70°C for 1 min, 65°C for 1 min, 60°C for 1 min, and 55°C for 1 min. Thirteen microliters of RT mix (4 μl 5× RT buffer, 1 μl 0.1 M DTT, 2 μl 10 mM dNTPs, 200 U SuperScript III, 20 U RNaseOUT [all from Invitrogen]) was added, and the reaction volume was incubated at 55°C for 1 h before inactivation (70°C for 15 min). For analysis, 9 μl Sequitherm Excel II stop/loading buffer (Epicentre) was added, incubated at 92°C for 3 min, and loaded onto an 8% urea polyacrylamide gel in a LiCor 4200 sequencer (LiCor Biosciences). As markers and ladders on the gels, 1-kb regions containing putative promoters were PCR amplified, TOPO TA cloned into pCR2.1 (Invitrogen), and sequenced using IRD800-labeled primers with the SequiTherm Excel II DNA sequencing kit (Epicentre).

Construction of GFP transcriptional fusions and in vitro fluorescence measurements.

Vector pAD123 with a promoterless red-shifted green fluorescent protein (GFP) gene (gfpmut3a) (11) was used for promoter studies. pAD43-25 with a constitutive promoter controlling gfpmut3a served as a positive control (15). In brief, putative ces promoter regions and sequentially deleted putative promoters were PCR amplified and directionally cloned in the multiple-cloning site of pAD123, giving rise to the pMKD plasmids shown in Table Table1.1. E. coli TOP10 (Invitrogen) was used for cloning steps, and correct ligation was confirmed by colony PCR and restriction analysis. The plasmids were passaged through the methylase-deficient E. coli strain INV110 (Invitrogen) and electroporated into B. cereus F4810/72 as described elsewhere (21). B. cereus main cultures bearing the recombinant gfpmut3a promoter probe vectors were inoculated with 100 μl of a 100-fold-diluted preculture and then cultured at 30°C and 150 rpm for 24 h. Cells were harvested by centrifugation (8,000 × g for 2 min at 4°C), washed once with phosphate-buffered saline, and collected by centrifugation. Fluorescence was measured as described previously (8). Fluorescence values (relative fluorescence units [RFU]) were normalized to the culture optical density at 600 nm (OD600) at the time of sampling. Background fluorescence from strains transformed with promoterless pAD123 was recorded.

Promoter fusion vectors

Assembly of the ces promoter-luciferase reporter system.

The promoterless luciferase luxABCDE vector pXen1 (27), optimized to study gene regulation in Gram-positive bacteria, was used as a basis for constructing a bioreporter system for monitoring ces toxin gene transcription. A 238-bp region of the main cereulide synthetase promoter was amplified by PCR from genomic DNA of Bacillus cereus F4810/72 by using the primers Pshort_for (5′-TAGAGAATTCCTGTTAGCCAATATAACGAGTC-3′) and Pshort_rev (5′-TTCCGGATCCTTCATTGAAAAATCCCTCT-3′) (EcoRI and BamHI restriction sites are underlined). This region corresponds to nucleotides (nt) 36891 to 37128 on the plasmid pCER270 (also designated pBCE) (GenBank accession number DQ889676 [52]). The fragment was cloned into the EcoRI and BamHI sites of the plasmid pXen1, creating pMDX[P1/luxABCDE], and further amplified in E. coli TOP10 (Invitrogen). The construct was then passaged through nonmethylating E. coli INV110 (Invitrogen) prior to transfer into B. cereus F4810/72 by electroporation, giving rise to F4810/72(pMDX[P1/luxABCDE]). To exclude cryptic promoter activity, B. cereus F4810/72 with a promoterless luciferase construct was used as a negative control. Furthermore, the luciferase reporter plasmid was used to transform two other emetic strains by electroporation: one clinical isolate and one food isolate from a recent emetic outbreak caused by a rice dish. All constructs were verified by restriction analysis and double-strand sequencing.

Inoculation of growth media.

Brain heart infusion (BHI; Merck), Trypticase soy agar (TSA; Merck), fortified nutrient agar (FNA; Oxoid), Columbia blood agar (Oxoid), and BCM Bacillus cereus group plating medium (Biosynth) were obtained from diverse distributors as prepared media, while the standard plate count (PC) medium, sodium lactate medium (27 ml of a 50% sodium lactate solution, 10 g yeast extract, 10 g peptone, 5 g KH2PO4 per liter), and mannitol egg yolk polymyxin agar (MYP) (1 g beef extract, 10 g peptone, 10 g mannitol, 10 g sodium chloride, 25 mg phenol red, 100 ml egg yolk solution, 100,000 IU polymyxin B supplement per liter) were prepared using standard chemicals and ingredients. All media were solidified with 15 g agar per liter. The media were inoculated with two 25-μl drops of overnight cultures of B. cereus F4810/72(pMDX[P1/luxABCDE]) and the negative control strain, B. cereus(pXen1[luxABCDE]), and incubated at 24°C for 24 h.

Inoculation of food samples.

Various foods were obtained from local consumer markets and prepared according to the manufacturer's instructions if necessary. Under sterile conditions, 30-g portions were filled in petri dishes and were spot inoculated with four 25-μl drops of B. cereus F4810/72 or B. cereus pMDX[P1/luxABCDE] overnight cultures. Plates were sealed to prevent moisture evaporation and were incubated for 24 h at 24°C. The initial cell inoculum per gram of food was determined by plating appropriate dilutions of the overnight culture on LB.

Growth of Bacillus cereus in food.

Spiked food samples (30 g) were comminuted for 1 min with 270 ml of a 0.025% Tween 80 solution (pH 7.0) using a stomacher. The homogenates were serially 10-fold diluted in LB, and 100 μl was spread in duplicate onto LB plates. For growth experiments with bioluminescent B. cereus, LB agar with 5 μg ml−1 chloramphenicol was used. Colonies showing the typical morphology of emetic Bacillus cereus were counted after 16 to 20 h of incubation at 30°C. The increase of viable cell counts was calculated as logarithmical growth units per gram of food.

Luciferase (lux) assay.

Luciferase activity of F4810/72 transformed with pMDX[P1/luxABCDE] was visualized on growth media and on food with a photon-counting intensified-charge-coupled-device (ICCD) camera (model 2400-32; Hamamatsu Photonics). Images were acquired for 2 s with a binning factor of 1 (without filter; relative aperture, 1), and the bioluminescence intensity was superimposed as false-color renderings. For quantification of the bioluminescence signals, region of interest (ROI) analysis was performed according to the manufacturer's instructions by using Living Image 2.10 software (Caliper Life Sciences) along with the IGOR Pro 4.01 software (WaveMetrics). In brief, the circle ROI tool was used to manually define the entire petri dish area as the ROI (dimensions were kept constant throughout all experiments), and the bioluminescence intensities was recorded as the total photon counts for all pixels inside the ROI (total counts). An empty area in the image was included as the average background ROI to correct for autofluorescence and was used for background correction of the signals.

Determination of cereulide contents in spiked food samples.

Whole portions of inoculated food were extracted with 20 ml of 96% ethanol for 24 h at room temperature on a rotary shaker. Extracts were centrifuged twice (9,000 × g for 20 min at 24°C). To estimate the toxin recovery rate, controls were spiked with valinomycin (Fluka) to a final concentration of 25 μg g−1, left for 1 h at room temperature, and extracted as described above. The HEp-2 cell-based bioassay was carried out as described previously (38). Additionally, extracts of control food samples were measured in order to exclude an interfering toxic background effect of the different ethanolic extracts on the HEp-2 cells. For calculation of the recovery rate of the toxins from food, samples were spiked with valinomycin and extracted as described above.


Polycistronic transcript of ces genes.

A complete transcriptional analysis using RT-PCR on the ces gene cluster showed that the cesPTABCD gene cluster forms a polycistronic operon. As RT-PCR primers were designed such that the amplicons overlap, the RT-PCR analysis covered the entire ces gene cluster (Fig. (Fig.1A).1A). A transcript was observed between the genes encoding the modules incorporating d-O-Leu and d-Ala (cesA) and l-O-Val and l-Val (cesB), and another transcript was observed in the noncoding regions between the CDSs for cesP, cesT, cesA, cesB, cesC, and cesD (Fig. (Fig.1B)1B) but not between cesH and cesP. PCR amplicons from the cDNA and DNA used as the positive control were the same size, whereas control reactions using RNA produced no amplicon. No transcript was detected by using a forward primer in the cesD coding region with a reverse primer located after a predicted hairpin downstream of cesD (Fig. (Fig.1B,1B, lane 10), indicating that this inverted repeat represents the terminator of ces gene transcription. Thus, cesPTABCD are transcribed as a single 23-kb transcript, while cesH is transcribed as monocistronic transcript from its own promoter. As bacteria regulate virulence factor expression not only at the level of transcription initiation but also by mRNA turnover rates, the decay rate of the mRNA of the main ces transcript was assessed to determine its stability. The ces transcript was relatively short-lived. Eighty percent of cesA mRNAs were degraded within 10 min (data not shown).

FIG. 1.
Transcriptional analysis of the ces gene cluster. (A) Bars indicate overlapping primer pairs used for transcriptional analysis of the ces operon (listed in Table S1 in the supplemental material). Bent arrows indicate promoters as determined by primer ...

Transcription initiation sites for the ces gene cluster mapped by RACE.

Transcription initiation sites were mapped using 5′ RACE (Fig. (Fig.2)2) and were confirmed by primer extension (data not shown). Analysis of RNA from liquid cultures grown to mid-log phase by using RACE revealed a transcription start site for cesH 76 bp upstream of the translational start site (corresponding to position 3266 in the ces sequence [GenBank DQ360825 {17}]) and two transcriptional start sites for cesP, located 100 bp upstream (P1) and 256 bp (P2) upstream of the translational start (Fig. (Fig.2).2). In addition, intracistronic promoters that might enhance transcription or be active under specific conditions were detected 56 bp upstream of cesT and 80 bp upstream of cesB (corresponding to position 6110 [cesT] and position 17248 [cesB] in the ces sequence [GenBank DQ360825]). In contrast, no intracistronic promoters were detected for cesA or cesC (neither by RACE nor by PE).

FIG. 2.
5′ RACE mapping of ces promoter sites. RACE products detected for the ces genes indicate intracistronic promoters present (cesT and cesB). Central promoters upstream of cesP (P1 and P2), indicated by an asterisk, and the cesP promoter region are ...

Transcriptional promoter fusions.

To investigate the active promoter via deletion analysis, the putative promoter regions were PCR amplified and cloned into pAD123 (15) upstream of the GFP gene gfpmut3a (11). An overview of the transcriptional fusions and their promoter activities is provided in Table Table1.1. The fusion of a DNA fragment designated P1 upstream of the cesP gene produced the strongest fluorescence signals, which were about 5- to 10-fold higher than the fluorescence signals from the constitutive promoter upp, used as the positive control. P2 alone was not active, and P0, a sequence downstream of P1, did not show activity, either. The cesH, cesT, and cesB promoters were only weakly active under the growth conditions tested. cesH promoter activity was in the same range as the positive control, while the intracistronic cesB promoter fragment showed about 50% of the fluorescence signal observed for the positive control. The cesT promoter fusion yielded a fluorescence signal of about 20% of the signal from the positive control. Therefore, P1 appears to be the main promoter active under these culture conditions.

Real-time monitoring of cereulide synthetase promoter activity by using a bioluminescent B. cereus reporter strain.

To monitor the activity of the main P1 promoter of the ces gene cluster in different environments, a 238-bp region of the cereulide synthetase promoter P1 was fused to a luciferase cassette on the pXen1 plasmid, creating pMDX[P1/luxABCDE] (see Materials and Methods for details). F4810/72 transformed with a promoterless luciferase construct was used as the negative control. After spot inoculation of media of diverse nutrient compositions, the reporter strain could be easily detected with the ICCD camera system, whereas luminescence was not emitted by the negative control strain (see Fig. S1 in supplemental material). The intensity of the luminescence signals was strongly dependent on the growth substrates available, indicating that transcription of the cereulide synthetase genes was promoted by media containing carbohydrates (MYP, PC medium, and BCM agar), while the P1 promoter activity was lower on media such as Columbia agar, BHI, and TSA, which contain larger amounts of rich proteinaceous ingredients (e.g., sheep blood, brain heart infusion, and peptones [≥20 g liter−1]). Media which promoted P1 activity also contained yeast or meat extracts (MYP, PC medium, LB, BCM, and FNA) and/or larger amounts of sodium chloride (MYP and LB).

Categorization of foods by cereulide production potential.

To evaluate the suitability of the luciferase reporter strain for analyzing the risk of cereulide formation in foods, a selection of food products with different nutrient compositions was inoculated with four 25-μl drops of a bioluminescent reporter strain culture and incubated at 24°C, mimicking temperature abuse. To further assess the influence of the luciferase reporter plasmid on cell multiplication, viable cells for both wild-type B. cereus F4810/72 and the bioluminescent derivative strain were enumerated after growth in different model food systems by conventional plate counting. Almost identical cell numbers were observed for the two strains after incubation for 24 h at 24°C (Table (Table2).2). To determine the activity of the P1 promoter in the artificially inoculated foods, software-assisted quantifications of the bioluminescence signals were performed using a Hamamatsu image collector and the Living Image software along with IGOR Pro 4.01 software (for details, see Materials and Methods). According to the transcription efficiency of the ces gene cluster, foods were categorized into three main classes depending on their potential to support cereulide production: high-risk, risk, and low-risk foods (Fig. (Fig.3A3A and Table Table2).2). To correlate the P1-driven transcript synthesis with the amount of cereulide produced by wild-type B. cereus F4810/72, the toxin quantity was determined with the HEp-2 assay. As illustrated in Table Table2,2, cereulide amounts correspond to the luminescence-defined risk categories.

FIG. 3.
Real-time monitoring of cereulide synthetase promoter activity in food. (A) The bioluminescent reporter strain B. cereus F4810/72(pMDX[P1/luxABCDE]) was inoculated into various foods. After incubation for 24 h at 24°C, the luciferase gene expression ...
Correlation of cereulide synthetase promoter activity, cell counts, and cereulide production of emetic B. cereus F4810/72 in different foods after 24 h at 24°Ca

To further evaluate the functionality of the luciferase-P1 construct in different genetic backgrounds, two other emetic B. cereus strains were grown on rice, and luminescence was recorded with the ICCD camera. As shown in Fig. Fig.3B,3B, P1 was also active in the clinical isolate as well as in a recently isolated emetic food-borne outbreak strain.


Central promoter drives polycistronic transcription of the Bacillus cereus cereulide toxin genes.

The transcriptional analysis of the ces gene locus presented here shows that cesH is transcribed from its own promoter, while the cesPTABCD genes are cotranscribed as a single large 23-kb polycistronic transcript (Fig. (Fig.1).1). A hairpin structure, predicted by Mfold and confirmed by real-time RT-PCR analysis (Fig. (Fig.1),1), directly downstream of cesD represents the terminator of the ces operon. Other large NRPS gene clusters are also expressed as polycistronic transcripts (see, e.g., references 12, 36, and 63), thus favoring coordinated expression of related genes. Due to the size of the NRP synthesis machinery and the wide range of unusual substrates used by NRPS, timing of transcription of NRPS genes is of special importance. Currently, over 350 nonproteinogenic amino acids are known, and most of them are activated and incorporated into microbial NRP metabolites (9). It would be energetically wasteful if the biosynthetic machineries required for NRP production were not tightly regulated and temporally coordinated. Surprisingly, the main ces promoter is located upstream of cesP and not cesA (Fig. (Fig.11 and and2),2), although 4′-PP transferase domains are highly conserved and not necessarily specific to their respective NPRS. In gramicidin S, which is produced by Bacillus brevis, the 4′-PP transferase gene (gsp) is located upstream of the structural grsTAB genes and transcribed from its own promoter Pgsp (36) yet can be used to complement in trans the deleted 4′-PP gene (sfp) required for surfactin biosynthesis (6). In the case of cereulide, it appears that the 4′-PP transferase encoded by cesP forms an integral part of the polycistronic ces gene cluster and the cesP promoter is essential for transcription of the ces gene cluster. Several +1 frameshifts occur in the ces operon, and although +1 frameshift mutations are less common than −1 frameshift mutations, they have been described for both prokaryotes and eukaryotes (23).

The function of internal promoters is unknown.

In addition to the central cesP promoter, intracistronic promoters for cesT and cesB, but not for cesA, were found. However, as shown by GFP promoter fusions, the internal promoters were only weakly active (Table (Table1).1). The occurrence of unusually long NRPS gene transcripts with multiple promoters and long leader sequences has been previously described for the mcy gene cluster responsible for microcystin biosynthesis in Microcystis aeruginosa (34); however, regulation of the intracistronic promoters and their impact on transcription are still cryptic. Intracistronic promoters, such as those found for cesT and cesB, might ensure adequate expression of distal genes and/or might be activated under particular growth conditions.

Bioluminescent reporter system for real-time monitoring of ces gene transcription.

In this study, the in situ application of the B. cereus F4810/72(pMDX[P1/luxABCDE]) reporter strain allowed a direct monitoring of ces promoter activity in complex environments. The results from various growth media and foods illustrated that the influence of environmental parameters is reflected by gradual promoter responses, which were displayed at the various intensities of bioluminescence signals (Table (Table22 and Fig. Fig.3A;3A; see also Fig. S1 in the supplemental material). The activity of the P1 promoter was strongly dependent on the nutrients provided. Thus, the lux reporter system described here visualizes the findings of previous studies reporting substantial variances in cereulide production on different growth media for a certain strain (50, 59). Media routinely used for B. cereus detection and experimental procedures (MYP, PC medium, LB, and BCM) led to higher P1 activities than protein-enriched media such as Columbia blood, TSA, and BHI agar (see Fig. S1 in the supplemental material). However, blood, TSA, and BHI agar are still frequently used to study cereulide production by emetic B. cereus, though it was reported earlier that less toxin was detected than with, e.g., skim milk- or rice-based media (26, 59). Correspondingly, high promoter activity was detected in the carbohydrate-rich cooked rice, while only intermediate promoter activity was observed in the proteinaceous Camembert cheese (Fig. (Fig.33 and Table Table2),2), although final cell numbers of the lux reporter strain in the two model food systems did not significantly differ. These findings correlate with reports of emetic food poisoning cases that implicate farinaceous rather than proteinaceous foods (3, 14, 18, 41) and underscore the risk of emetic intoxications, for instance, in mass catering quantities of rice- or pasta-based dishes that have been improperly cooled.

B. cereus F4810/72 was originally isolated from rice involved in an emetic food poisoning case in the United Kingdom in 1972 (61). As decades of laboratory culturing might have modified the original features of this strain, two other recently obtained emetic isolates were transformed with the luciferase reporter vector and included in this study (Fig. (Fig.3B).3B). The responses of the main cereulide promoters in the background of a clinical isolate and of an emetic outbreak isolate did not differ discernibly on cooked rice, indicating that F4810/72 is an appropriate tool for real-time monitoring of ces gene expression in different environments.

Categorization of foods by potential for cereulide production.

A selection of retail food products was inoculated with a bioluminescent emetic B. cereus reporter strain and incubated at 24°C, the temperature reported to be optimal for cereulide production (31). Under these conditions, the foods could be divided into three main categories by their potential to support toxin synthesis (Fig. (Fig.3A3A and Table Table2).2). As shown by cereulide quantification, the amount of toxin produced correlated to the signal intensity of the ces promoter-driven lux gene expression (Table (Table2).2). Several rice dishes were implicated in food poisoning cases or previously reported to support toxin synthesis to a great extent (e.g., see references 25, 28, and 35). Consistently, cooked rice was classified as a high-risk food in our system. The system revealed other products besides this classical outbreak-related food to be potentially at risk, such as a sweetened dairy-based dessert and Camembert cheese. Interestingly, cereulide was detected in another Camembert cheese sample investigated previously (49). The proposed risk group classification is further supported by the fact that around 8 to 10 μg cereulide per kg of body weight must be consumed by healthy adults to provoke a clinical manifestation of the emetic syndrome (33, 56). Thus, foods containing >1 μg g−1 cereulide would be hazardous to a child with a body weight of 10 kg if about 100 g were consumed. Indeed, foods related to emetic poisoning contained about 1 to 3 μg g−1 cereulide (3, 33). But the risk-group foods should also be considered critical, as other studies revealed that food poisoning cases implicated meals which contained as little as 0.01 μg g−1 cereulide (3).

A comprehensive evaluation concerning cereulide production in different groups of food is still missing, most probably due to the time-consuming and laborious methods required for toxin quantification in these complex matrices. The extraction of the highly lipophilic cereulide molecule, especially from fat-rich foods, is often prone to error (for example, the toxin recovery rate from different food matrices varies from 22% to 100% [M. Ehling-Schulz and E. Frenzel, unpublished data]), and matrix-specific extraction protocols are often required (55). The assay presented here offers the advantage of a high-throughput system for a basic categorization of foods and food ingredients by the risk of cereulide production. Thus, the real-time monitoring of ces gene expression might have the potential as an industrial application to support hazard identification in terms of hazard analysis and critical control point concepts and offers the possibility of basic identification of a potential source of risk to the consumer.

This study focuses on the classification of foods by their potential to support cereulide toxin synthesis. However, it should also be taken into consideration that B. cereus is able to produce additional food spoilage and virulence factors, including phospholipases, proteases, hemolysins, and enterotoxins, which are unambiguously of medical and economical impact (10, 35, 54). lux gene reporter systems, similar to the one described in this work, might also be useful for monitoring enterotoxin expression and could add important information to the hazard identification of different food systems.

Supplementary Material

[Supplemental material]


We thank Daniel Zeigler of the Bacillus Genetic Stock Center at the Ohio State University and Jo Handelsman of the Department of Plant Pathology at the University of Wisconsin—Madison for the kind gifts of pAD123 and pAD43-25, respectively. We thank Kevin Francis (Xenogen Corporation) for the gift of pXen1 and Romy Renner and Laura Tschernek for excellent technical assistance.

This work was supported in part by the Forschungskreis der Ernährungsindustrie e.V. (FEI), by the Arbeitskreis für Industrielle Forschung (AiF), by the Ministry of Economics and Technology (BMWi; project no. 15186N), and in part by the Bavarian Ministry for Agriculture and Forestry (project no. M2-7606.2-526).


[down-pointing small open triangle]Published ahead of print on 28 December 2009.

Supplemental material for this article may be found at


1. Agaisse, H., M. Gominet, O. A. Okstad, A.-B. Kolsto, and D. Lereclus. 1999. PlcR is a pleiotropic regulator of extracellular virulence factor gene expression in Bacillus thuringiensis. Mol. Microbiol. 32:1043-1053. [PubMed]
2. Agata, N., M. Mori, M. Ohta, S. Suwan, I. Ohtani, and M. Isobe. 1994. A novel dodecadepsipeptide, cereulide, isolated from Bacillus cereus causes vacuole formation in HEp-2 cells. FEMS Microbiol. Lett. 121:31-34. [PubMed]
3. Agata, N., M. Ohta, and K. Yokoyama. 2002. Production of Bacillus cereus emetic toxin (cereulide) in various foods. Int. J. Food Microbiol. 73:23-27. [PubMed]
4. Beecher, D. J., and J. D. Macmillan. 1991. Characterization of the components of hemolysin BL from Bacillus cereus. Infect. Immun. 59:1778-1784. [PMC free article] [PubMed]
5. Bertani, G. 1951. Studies on lysogenesis. I. The mode of phage liberation by lysogenic Escherichia coli. J. Bacteriol. 62:293-300. [PMC free article] [PubMed]
6. Borchert, S., T. Stachelhaus, and M. Marahiel. 1994. Induction of surfactin production in Bacillus subtilis by gsp, a gene located upstream of the gramicidin S operon in Bacillus brevis. J. Bacteriol. 176:2458-2462. [PMC free article] [PubMed]
7. Brendel, V., and E. N. Trifonov. 1984. A computer algorithm for testing potential prokaryotic terminators. Nucleic Acids Res. 12:4411-4427. [PMC free article] [PubMed]
8. Bresolin, G., K. Neuhaus, S. Scherer, and T. M. Fuchs. 2006. Transcriptional analysis of long-term adaptation of Yersinia enterocolitica to low-temperature growth. J. Bacteriol. 188:2945-2958. [PMC free article] [PubMed]
9. Caboche, S., M. Pupin, V. Leclere, A. Fontaine, P. Jacques, and G. Kucherov. 2008. NORINE: a database of nonribosomal peptides. Nucleic Acids Res. 36:D326-D331. [PMC free article] [PubMed]
10. Callegan, M. C., S. T. Kane, D. C. Cochran, B. Novosad, M. S. Gilmore, M. Gominet, and D. Lereclus. 2005. Bacillus endophthalmitis: roles of bacterial toxins and motility during infection. Invest. Ophthalmol. Vis. Sci. 46:3233-3238. [PubMed]
11. Cormack, B. P., R. H. Valdivia, and S. Falkow. 1996. FACS-optimized mutants of the green fluorescent protein (GFP). Gene 173:33-38. [PubMed]
12. Cosmina, P., F. Rodriguez, F. de Ferra, G. Grandi, M. Perego, G. Venema, and D. van Sinderen. 1993. Sequence and analysis of the genetic locus responsible for surfactin synthesis in Bacillus subtilis. Mol. Microbiol. 8:821-831. [PubMed]
13. Crosa, J. H., and C. T. Walsh. 2002. Genetics and assembly line enzymology of siderophore biosynthesis in bacteria. Microbiol. Mol. Biol. Rev. 66:223-249. [PMC free article] [PubMed]
14. Dierick, K., E. Van Coillie, I. Swiecicka, G. Meyfroidt, H. Devlieger, A. Meulemans, G. Hoedemaekers, L. Fourie, M. Heyndrickx, and J. Mahillon. 2005. Fatal family outbreak of Bacillus cereus-associated food poisoning. J. Clin. Microbiol. 43:4277-4279. [PMC free article] [PubMed]
15. Dunn, A. K., and J. Handelsman. 1999. A vector for promoter trapping in Bacillus cereus. Gene 226:297-305. [PubMed]
16. Duport, C., A. Zigha, E. Rosenfeld, and P. Schmitt. 2006. Control of enterotoxin gene expression in Bacillus cereus F4430/73 involves the redox-sensitive ResDE signal transduction system. J. Bacteriol. 188:6640-6651. [PMC free article] [PubMed]
17. Ehling-Schulz, M., M. Fricker, H. Grallert, P. Rieck, M. Wagner, and S. Scherer. 2006. Cereulide synthetase gene cluster from emetic Bacillus cereus: structure and location on a mega virulence plasmid related to Bacillus anthracis toxin plasmid pXO1. BMC Microbiol. 6:20. [PMC free article] [PubMed]
18. Ehling-Schulz, M., M. Fricker, and S. Scherer. 2004. Bacillus cereus, the causative agent of an emetic type of food-borne illness. Mol. Nutr. Food Res. 48:479-487. [PubMed]
19. Ehling-Schulz, M., M. Fricker, and S. Scherer. 2004. Identification of emetic toxin producing Bacillus cereus strains by a novel molecular assay. FEMS Microbiol. Lett. 232:189-195. [PubMed]
20. Ehling-Schulz, M., B. Svensson, M.-H. Guinebretiere, T. Lindback, M. Andersson, A. Schulz, M. Fricker, A. Christiansson, P. E. Granum, E. Martlbauer, C. Nguyen-The, M. Salkinoja-Salonen, and S. Scherer. 2005. Emetic toxin formation of Bacillus cereus is restricted to a single evolutionary lineage of closely related strains. Microbiology 151:183-197. [PubMed]
21. Ehling-Schulz, M., N. Vukov, A. Schulz, R. Shaheen, M. Andersson, E. Martlbauer, and S. Scherer. 2005. Identification and partial characterization of the nonribosomal peptide synthetase gene responsible for cereulide production in emetic Bacillus cereus. Appl. Environ. Microbiol. 71:105-113. [PMC free article] [PubMed]
22. Fagerlund, A., O. Ween, T. Lund, S. P. Hardy, and P. E. Granum. 2004. Genetic and functional analysis of the cytK family of genes in Bacillus cereus. Microbiology 150:2689-2697. [PubMed]
23. Farabaugh, P. J. 1996. Programmed translational frameshifting. Annu. Rev. Genet. 30:507-528. [PubMed]
24. Finking, R., and M. A. Marahiel. 2004. Biosynthesis of nonribosomal peptides. Annu. Rev. Microbiol. 58:453-488. [PubMed]
25. Finlay, W. J., N. A. Logan, and A. D. Sutherland. 2002. Bacillus cereus emetic toxin production in cooked rice. Food Microbiol. 19:431-439.
26. Finlay, W. J., N. A. Logan, and A. D. Sutherland. 1999. Semiautomated metabolic staining assay for Bacillus cereus emetic toxin. Appl. Environ. Microbiol. 65:1811-1812. [PMC free article] [PubMed]
27. Francis, K. P., D. Joh, C. Bellinger-Kawahara, M. J. Hawkinson, T. F. Purchio, and P. R. Contag. 2000. Monitoring bioluminescent Staphylococcus aureus infections in living mice using a novel luxABCDE construct. Infect. Immun. 68:3594-3600. [PMC free article] [PubMed]
28. Fricker, M., U. Messelhausser, U. Busch, S. Scherer, and M. Ehling-Schulz. 2007. Diagnostic real-time PCR assays for the detection of emetic Bacillus cereus strains in foods and recent food-borne outbreaks. Appl. Environ. Microbiol. 73:1892-1898. [PMC free article] [PubMed]
29. Gohar, M., K. Faegri, S. Perchat, S. Ravnum, O. A. Okstad, M. Gominet, A. B. Kolsto, and D. Lereclus. 2008. The PlcR virulence regulon of Bacillus cereus. PLoS One 3:e2793. [PMC free article] [PubMed]
30. Granum, P. E., K. O'Sullivan, and T. Lund. 1999. The sequence of the non-haemolytic enterotoxin operon from Bacillus cereus. FEMS Microbiol. Lett. 177:225-229. [PubMed]
31. Häggblom, M. M., C. Apetroaie, M. A. Andersson, and M. S. Salkinoja-Salonen. 2002. Quantitative analysis of cereulide, the emetic toxin of Bacillus cereus, produced under various conditions. Appl. Environ. Microbiol. 68:2479-2483. [PMC free article] [PubMed]
32. Ishii, T., K.-I. Yoshida, G. Terai, Y. Fujita, and K. Nakai. 2001. DBTBS: a database of Bacillus subtilis promoters and transcription factors. Nucleic Acids Res. 29:278-280. [PMC free article] [PubMed]
33. Jääskeläinen, E. L., V. Teplova, M. A. Andersson, L. C. Andersson, P. Tammela, M. C. Andersson, T. I. Pirhonen, N.-E. L. Saris, P. Vuorela, and M. S. Salkinoja-Salonen. 2003. In vitro assay for human toxicity of cereulide, the emetic mitochondrial toxin produced by food poisoning Bacillus cereus. Toxicol. In Vitro 17:737-744. [PubMed]
34. Kaebernick, M., E. Dittmann, T. Borner, and B. A. Neilan. 2002. Multiple alternate transcripts direct the biosynthesis of microcystin, a cyanobacterial nonribosomal peptide. Appl. Environ. Microbiol. 68:449-455. [PMC free article] [PubMed]
35. Kramer, J. M., and R. J. Gilbert. 1989. Bacillus cereus and other Bacillus species, p. 21-70. In M. P. Doyle (ed.), Food-borne bacterial pathogens. Marcel Dekker, New York, NY.
36. Krätzschmar, J., M. Krause, and M. A. Marahiel. 1989. Gramicidin S biosynthesis operon containing the structural genes grsA and grsB has an open reading frame encoding a protein homologous to fatty acid thioesterases. J. Bacteriol. 171:5422-5429. [PMC free article] [PubMed]
37. Logemann, J., J. Schell, and L. Willmitzer. 1987. Improved method for the isolation of RNA from plant tissues. Anal. Biochem. 163:16-20. [PubMed]
38. Lücking, G., M. K. Dommel, S. Scherer, A. Fouet, and M. Ehling-Schulz. 2009. Cereulide synthesis in emetic Bacillus cereus is controlled by the transition state regulator AbrB, but not by the virulence regulator PlcR. Microbiology 155:922-931. [PubMed]
39. Lund, T., M. De Buyser, and P. Granum. 2000. A new cytotoxin from Bacillus cereus that may cause necrotic enteritis. Mol. Microbiol. 38:254-261. [PubMed]
40. Magarvey, N. A., M. Ehling-Schulz, and C. T. Walsh. 2006. Characterization of the cereulide NRPS hydroxy acid specifying modules: activation of keto acids and chiral reduction on the assembly line. J. Am. Chem. Soc. 128:10698-10699. [PubMed]
41. Mahler, H., A. Pasi, J. M. Kramer, P. Schulte, A. C. Scoging, W. Bar, and S. Krahenbuhl. 1997. Fulminant liver failure in association with the emetic toxin of Bacillus cereus. N. Engl. J. Med. 336:1142-1148. [PubMed]
42. Marahiel, M. A., T. Stachelhaus, and H. D. Mootz. 1997. Modular peptide synthetases involved in nonribosomal peptide synthesis. Chem. Rev. 97:2651-2674. [PubMed]
43. Martineau, F., F. J. Picard, P. H. Roy, M. Ouellette, and M. G. Bergeron. 1996. Species-specific and ubiquitous DNA-based assays for rapid identification of Staphylococcus epidermidis. J. Clin. Microbiol. 34:2888-2893. [PMC free article] [PubMed]
44. Moon, S., Y. Byun, H.-J. Kim, S. Jeong, and K. Han. 2004. Predicting genes expressed via −1 and +1 frameshifts. Nucleic Acids Res. 32:4884-4892. [PMC free article] [PubMed]
45. Okstad, O. A., M. Gominet, B. Purnelle, M. Rose, D. Lereclus, and A.-B. Kolsto. 1999. Sequence analysis of three Bacillus cereus loci carrying PlcR-regulated genes encoding degradative enzymes and enterotoxin. Microbiology 145:3129-3138. [PubMed]
46. Ouhib, O., T. Clavel, and P. Schmitt. 2006. The production of Bacillus cereus enterotoxins is influenced by carbohydrate and growth rate. Curr. Microbiol. 53:222-226. [PubMed]
47. Pfaffl, M. W. 2001. A new mathematical model for relative quantification in real-time RT-PCR. Nucleic Acids Res. 29:e45. [PMC free article] [PubMed]
48. Pósfay-Barbe, K. M., J. Schrenzel, J. Frey, R. Studer, C. Korff, D. C. Belli, P. Parvex, P. C. Rimensberger, and M. G. Schappi. 2008. Food poisoning as a cause of acute liver failure. Pediatr. Infect. Dis. J. 27:846-847. [PubMed]
49. Rajkovic, A., M. Uyttendaele, and J. Debevere. 2007. Computer aided boar semen motility analysis for cereulide detection in different food matrices. Int. J. Food Microbiol. 114:92-99. [PubMed]
50. Rajkovic, A., M. Uyttendaele, W. Deley, A. Van Soom, T. Rijsselaere, and J. Debevere. 2006. Dynamics of boar semen motility inhibition as a semi-quantitative measurement of Bacillus cereus emetic toxin (cereulide). J. Microbiol. Methods 65:525-534. [PubMed]
51. Rajkovic, A., M. Uyttendaele, A. Vermeulen, M. Andjelkovic, I. Fitz-James, P. in 't Veld, Q. Denon, R. Verhe, and J. Debevere. 2008. Heat resistance of Bacillus cereus emetic toxin, cereulide. Lett. Appl. Microbiol. 46:536-541. [PubMed]
52. Rasko, D. A., M. J. Rosovitz, O. A. Okstad, D. E. Fouts, L. Jiang, R. Z. Cer, A.-B. Kolsto, S. R. Gill, and J. Ravel. 2007. Complete sequence analysis of novel plasmids from emetic and periodontal Bacillus cereus isolates reveals a common evolutionary history among the B. cereus group plasmids including B. anthracis pXO1. J. Bacteriol. 189:52-64. [PMC free article] [PubMed]
53. Reese, M. G. 2001. Application of a time-delay neural network to promoter annotation in the Drosophila melanogaster genome. Comp. Chem. 26:51-56. [PubMed]
54. Schoeni, J. L., and A. C. L. Wong. 2005. Bacillus cereus food poisoning and its toxins. J. Food Prot. 68:636-648. [PubMed]
55. Shaheen, R., M. A. Andersson, C. Apetroaie, A. Schulz, M. Ehling-Schulz, V. M. Ollilainen, and M. S. Salkinoja-Salonen. 2006. Potential of selected infant food formulas for production of Bacillus cereus emetic toxin, cereulide. Int. J. Food Microbiol. 107:287-294. [PubMed]
56. Shinagawa, K., H. Konuma, H. Sekita, and S. Sugii. 1995. Emesis of rhesus monkeys induced by intragastric administration with the HEp-2 vacuolation factor (cereulide) produced by Bacillus cereus. FEMS Microbiol. Lett. 130:87-90. [PubMed]
57. Sieber, S. A., and M. A. Marahiel. 2005. Molecular mechanisms underlying nonribosomal peptide synthesis: approaches to new antibiotics. Chem. Rev. 105:715-738. [PubMed]
58. Stenfors Arnesen, L. P., A. Fagerlund, and P. E. Granum. 2008. From soil to gut: Bacillus cereus and its food poisoning toxins. FEMS Microbiol. Rev. 32:579-606. [PubMed]
59. Szabo, R., J. Speirs, and M. Akhtar. 1991. Cell culture detection and conditions for production of a Bacillus cereus heat-stable toxin. J. Food Prot. 54:272-276.
60. Thorsen, L., B. M. Hansen, K. F. Nielsen, N. B. Hendriksen, R. K. Phipps, and B. B. Budde. 2006. Characterization of emetic Bacillus weihenstephanensis, a new cereulide-producing bacterium. Appl. Environ. Microbiol. 72:5118-5121. [PMC free article] [PubMed]
61. Turnbull, P. C., J. M. Kramer, K. Jorgensen, R. J. Gilbert, and J. Melling. 1979. Properties and production characteristics of vomiting, diarrheal, and necrotizing toxins of Bacillus cereus. Am. J. Clin. Nutr. 32:219-228. [PubMed]
62. Vassileva, M., K. Torii, M. Oshimoto, A. Okamoto, N. Agata, K. Yamada, T. Hasegawa, and M. Ohta. 2007. A new phylogenetic cluster of cereulide-producing Bacillus cereus strains. J. Clin. Microbiol. 45:1274-1277. [PMC free article] [PubMed]
63. Wang, N., S.-E. Lu, Q. Yang, S.-H. Sze, and D. C. Gross. 2006. Identification of the syr-syp box in the promoter regions of genes dedicated to syringomycin and syringopeptin production by Pseudomonas syringae pv. syringae B301D. J. Bacteriol. 188:160-168. [PMC free article] [PubMed]
64. Zuker, M. 2003. Mfold web server for nucleic acid folding and hybridization prediction. Nucleic Acids Res. 31:3406-3415. [PMC free article] [PubMed]

Articles from Applied and Environmental Microbiology are provided here courtesy of American Society for Microbiology (ASM)