lable at ScienceDirect

Food Microbiology 55 (2016) 73e85

Contents lists avai

Food Microbiology

journal homepage: www.elsevier .com/locate/ fm

Transcriptome analysis of Bacillus thuringiensis spore life, germinationand cell outgrowth in a vegetable-based food model

Daniela Bassi a, Francesca Colla a, 1, Simona Gazzola a, Edoardo Puglisi a,Massimo Delledonne b, Pier Sandro Cocconcelli a, *

a Istituto di Microbiologia e Centro Ricerche Biotecnologiche, Universit�a Cattolica del Sacro Cuore, via Emilia Parmense 84, 29122 Piacenza, Italy e viaMilano 24, 26100 Cremona, Italyb Dipartimento di Biotecnologie, Universit�a degli Studi di Verona, Strada le Grazie 15, 37134 Verona, Italy

a r t i c l e i n f o

Article history:Received 23 April 2015Received in revised form3 November 2015Accepted 10 November 2015Available online 12 November 2015

Keywords:Bacillus cereus sensu latoFood modelMicroarraySporesGene expression

* Corresponding author.E-mail address: pier.cocconcelli@unicatt.it (P.S. Co

1 Current address. Colla S.p.a., Via Sant'Anna 10, 29

http://dx.doi.org/10.1016/j.fm.2015.11.0060740-0020/© 2015 Elsevier Ltd. All rights reserved.

a b s t r a c t

Toxigenic species belonging to Bacillus cereus sensu lato, including Bacillus thuringiensis, cause foodborneoutbreaks thanks to their capacity to survive as spores and to grow in food matrixes. The goal of thiswork was to assess by means of a genome-wide transcriptional assay, in the food isolate B. thuringiensisUC10070, the gene expression behind the process of spore germination and consequent outgrowth in avegetable-based food model. Scanning electron microscopy and Energy Dispersive X-ray microanalysiswere applied to select the key steps of B. thuringiensis UC10070 cell cycle to be analyzed with DNA-microarrays. At only 40 min from heat activation, germination started rapidly and in less than twohours spores transformed in active growing cells. A total of 1646 genes were found to be differentiallyexpressed and modulated during the entire B. cereus life cycle in the food model, with most of thesignificant genes belonging to transport, transcriptional regulation and protein synthesis, cell wall andmotility and DNA repair groups. Gene expression studies revealed that toxin-coding genes nheC, cytK andhblC were found to be expressed in vegetative cells growing in the food model.

© 2015 Elsevier Ltd. All rights reserved.

1. Introduction

The genus Bacillus comprises Gram-positive species that can befound ubiquitously in the environment and can survive and persistto starvation and adverse growth conditions thanks to their abilityto form resistant and metabolically inert spores (Schmidt et al.,2011). Six different Bacillus species, namely Bacillus anthracis, Ba-cillus cereus, Bacillus mycoides, Bacillus pseudomycoides, Bacillusthuringiensis and Bacillus weihenstephanensis, although exhibitingdifferent biological characteristics, such as pathogenicity and toxinproduction, have been classified on the basis of 16S sequencing(Daffonchio et al., 2003), multilocus sequence typing (MLST) (Priestet al., 2004) and comparative genome sequencing (Rasko et al.,2005) as a single taxonomical unit, the Bacillus cereus sensu lato(Helgason et al., 2000; Schmidt et al., 2011). Moreover, the B. cereussensu lato was recently extended to comprise two further species:Bacillus toyonensis, a strain employed in animal nutrition (Jim�enez

cconcelli).010 Cadeo, PC, Italy.

et al., 2013) and Bacillus cytotoxicus, which is a thermotolerantspecies associated with food poisonings (Guinebreti�ere et al., 2013).All these species are associated by their genome structure, but theyoften demonstrate specific traits that are due to large plasmid de-terminants (Jensen et al., 2003). Cry genes located on a largeplasmid in B. thuringiensis (Schnepf et al., 1998), cereulide gene on amega virulence plasmid in B. cereus (Hoton et al., 2005) and capsulegenes on pXO1 and pXO2 plasmids in B. anthracis (Ehling-Schulzet al., 2006) are only some examples. Moreover, the toxicologicalpotential, the occurrence of B. cereus-related food-borne outbreaks(Bargabus et al., 2002), the possible food contamination ofB. thuringiensis as a result of its wide biopesticide use (Hendriksenand Hansen, 2006), and the high level of genomic similarities be-tween the species forming the group (Schmidt et al., 2011), indicatethat the B. cereus sensu lato should be studied as a whole.

Bacillus spores are indeed often present in food matrices, wherethey can resist preservation treatments and germinate in the end-product if nutrients and proper conditions to return to vegetativelife are available (Dworkin and Shah, 2010). Their presence isindeed responsible for serious food-borne illnesses (EFSA, 2015)and significant numbers of food spoilage cases (Drean et al., 2015;

D. Bassi et al. / Food Microbiology 55 (2016) 73e8574

Markland et al., 2013). Factors triggering germination in food can berepresented by amino acids, sugars and other molecules or condi-tions (pH, sub-lethal thermal treatments, aw, storage conditions)that let the spore exit from its dormant state (Moir and Smith,1990;Moir et al., 2002), the precondition for the subsequent growth andtoxins production.

The spore inert state has been demonstrated with the evidenceof a jelly core and a partial dehydration that forms an immobilematrix (Cowan et al., 2003; Driks, 2002; Nakashio and Gerhardt,1985), but it's still disputed if the spore maintains some minimalmetabolic activity. What has been already demonstrated is thatspores of different Bacillus and Clostridium species contain RNAtranscripts (Bassi et al., 2013; Bettegowda et al., 2006; Dembeket al., 2013; Keijser et al., 2007) and that the amount of RNAseems to be affected by the spore incubation temperature andspore age (Segev et al., 2012). The function of these transcripts isstill largely unknown: they could represent preserved RNA mole-cules ready for a new cycle, or degraded molecules useful to be thestarting point for a de novo synthesis. The RNA state appears toaffect the exit from dormancy together with the effects of envi-ronmental signals (Segev et al., 2012). The Bacillus spores behaviourin a food matrix in terms of RNA state and gene expression duringgermination is also still unknown. Factors triggering germinationand outgrowth have been largely studied in Bacillus subtilis (Keijseret al., 2007) and B. cereus (Abee et al., 2011), while germinationstudies of B. cereus in foodstuffs have been recently performed inmilk (Bartoszewicz et al., 2013), cream, b�echamel sauce and mixedvegetable soup (Samapundo et al., 2014) and as a probabilisticmodel of predictive microbiology in refrigerated processed foods ofextended durability (Daelman et al., 2013). The germination pro-cess has been studied also in terms of gene expression by DNA-microarrays particularly in B. cereus for screening purposes(Sergeev et al., 2006), and to assess the microorganism' response tosalt stress (Den Besten et al., 2009), preservatives such as sorbicacid (VanMelis et al., 2011), disinfectants (Ceragioli et al., 2010) andmodified atmosphere (Passalacqua et al., 2009). DNA-microarrayswere also applied to assess the germination of spores in B. subtilis(Berka et al., 2002; Keijser et al., 2007), Clostridium novyi(Bettegowda et al., 2006), Clostridium sporogenes (Bassi et al., 2013)and Clostridium difficile (Dembek et al., 2013).

The aim of this work was to simulate a B. cereus contaminationin an experimental vegetable food model and to study the geneexpression profile during the initial colonization of food as sporeentities and during germination and outgrowth phases. To do this,we investigated the phenotypic behaviour of a strain isolated fromspoilt food through OD measurements, scanning electron micro-scopy (SEM) and Energy Dispersive X-ray (EDAX) microanalysisthat allowed us to determine the critical steps representing theentire time-course of Bacillus life cycle in a foodstuff. An array-based transcriptome analysis was then applied to study genesdifferentially expressed in these steps, particularly during thegermination process.

2. Materials and methods

2.1. Bacterial strain isolation, growth conditions and sporeproduction

The strain UC10070 was isolated from a biofilm on a spoiltvegetable-based puree using a B. cereus selective agar medium(Oxoid, Milan Italy). It was assigned to the Bacillus cereus sensu latoby means of 16S rRNA gene analysis, and specifically toB. thuringiensis on the basis of cry toxins identification. The strainwas cultured in Brain Heart Infusion (BHI) broth at 37 �C in aerobicconditions on continuous shaking.

Spores were produced from cells cultured in BP medium (Ba-cillus Genetic Stock Center, Ohio State University, Columbus, OH,USA). BP plates were inoculated with 500 ml of B. thuringiensisUC10070 overnight cultures, and incubated for 4 days at 37 �C.Spores were harvested and purified by extensive washing withdistilled water at 4 �C (Nicholson and Setlow, 1990). The sporecrops, checked by phase-contrast microscopy, were free (<1%) ofvegetative cells and germinating spores. Spore suspensions weremaintained at 4 �C, and immediately used for the subsequentanalyses.

2.2. Toxin assays

The presence of enterotoxin genes and their expression in B.thuringiensis UC10070 were assessed by means of PCR and RT-PCRanalyses. Primer sets and conditions listed in Table 1 were used todetect nheBC, hblCD, cytK, cry and ces genes.

The ability of the strain to produce the diarrhoeal enterotoxinHBL (haemolytic fraction L2) was also tested by the reverse passivelatex agglutination test using BCET-RPLA toxin detection kit(Oxoid). After 1% inoculum of the strain in BHI broth, and incuba-tion at 37 �C for 18 h on shaking (250 cycles/min), the culture wascentrifuged; supernatants were filter-sterilized and storedat �20 �C until the assay performance according to the manufac-turers' instructions.

2.3. Food model and germination assays

Three commercial thermally treated vegetable creams basedrespectively on pepper, artichoke and spinach, and an experimentalpasteurized soup, made with courgettes, potatoes and milk (CPM),were tested in triplicates. CPM food model was prepared by mixinghomogeneously fresh courgettes and potatoes, previously washed,trimmed and peeled, to UHT milk with a ratio of 3:1:1. The mixtureobtained was sterilized at 121 �C for 15 min. pH and aw values ofeach soup were determined before proceeding with the experi-ments. The samples were inoculated with 105 CFU/ml and incu-bated at 20 �C and at 4 �C for seven days. Growth dynamics forB. thuringiensis UC10070 were analyzed by plate counts.

Germination assays were carried out at room temperature byinoculating 100 ml of B. thuringiensis UC10070 spore suspension(107 CFU/ml) in selected food products followed by anaerobicpackaging with Anaerocult A packs (Merck, Darmstadt, Germany)and by a heat treatment for 15 min at 80 �C mimicking processingconditions.

The germination process in the CPM food model was monitoredby optical and scanning electron microscopy (SEM) together withEnergy Dispersive X-ray (EDAX) analyses as described below. Mi-croscopy data were confirmed by CFU counts plating before andafter pasteurization for 20 min at 80 �C respectively for total cells(vegetative and spores) and spore counts.

For expression studies, the time points T0 (dormant spores),40 min (GSP, germinating spores), 2 h (C2h, early-log phase) and12 h (C12h, mid-log phase) after thermal treatment, were selectedas the most representative of the B. thuringiensis UC10070 life cycleunder the tested conditions. Samples used for microscopy andmolecular analyses were taken from the inoculated CPM foodmodel at the above time points and immediately stored at �80 �Cfor RNA extraction. All studies were carried out in three indepen-dent experiments.

2.4. Microscopy examinations

The transition from spores to vegetative cells was monitored byoptical phase contrast microscopy, SEM and EDAX microanalyses.

Table 1Primer set for detection of enterotoxin genes in B. thuringiensis UC10070 isolated strain. For HBL and NHE complex, PCR analysis were performed to amplify at least two genesin the three component operons: hblC and hblD genes for the haemolytic toxin, nheB and nheC genes for the non-haemolytic toxin.

Target genes Primer name Sequence (50e30) Reference

cryIV DiplA CAAGCCGCAAATCTTGTGGA (Carozzi et al., 1991)Dip1B ATGGCTTGTTTCGCTACATC

cry1Ia1 X62821 Fw GCTGTCTACCATGATTCGCTTG (Song et al., 2003)X62821 Rv CAGTGCAGTAACCTTCTCTTGCA

hblC L2A Fw AATGGTCATCGGAACTCTAT (Hansen and Hendriksen, 2001)L2B Rv CTCGCTGTTCTGCTGTTAAT

hblD HBLD-N AATCAAGAGCTGGTCACGAAT (Hansen and Hendriksen, 2001)HBLD-C CACCAATTGACCATGCTAAT

nheB nheB 1500 S Fw CTATCAGCACTTATGGCAG (Hansen and Hendriksen, 2001)nheB 2269 A Rv ACTCCTAGCGGTGTTCC

nheC nheC 2820 S Fw CGGTAGTGATTGCTGGG (Hansen and Hendriksen, 2001)nheC 3401 A Rv CAGCATTCGTACTTGCCAA

cytK cytKf GATAATATGACAATGTCTTTAAA (Swiecicka and Mahillon, 2006)cytKr GGAGAGAAACCGCTATTTGT

ces CesF1 GGTGACACATTATCATATAAGGTG (Ehling-Schulz et al., 2005)CesR2 GTAAGCGAACCTGTCTGTAACAACA

rrn P1 GCGGCGTGCCTAATACATGC (Klijn et al., 1991)P4 ATCTACGCATTTCACCGCTAC

D. Bassi et al. / Food Microbiology 55 (2016) 73e85 75

Optical microscopy was made at regular time points (every 10 min)observing at least 15 optical fields for the presence of phase-brightspores, dark spores and vegetative cells. Samples for SEM obser-vations and EDAX were prepared according to the carbon coatingmethod previously described (Bassi et al., 2009). SEM and EDAXanalyses were carried out on three replicates for each analyzedcondition with a Philips XL30 E-SEM microscope.

2.5. RNA isolation

RNA was isolated from CPM frozen samples collected at theabove-defined time points and immediately frozen at�80 �C (threebiological replicates for each analyzed stage). RNA was isolatedfrom 1 ml of soup for each phase, homogenized with physiologicalsolution, centrifuged at 10,000 g for 10 min and pellets processedusing RNeasy mini kit (Qiagen, Ilden, Germany) according to themanufacturer's instructions. Samples lysis was carried out in aFastPrep instrument (MP Biomedicals, Irvine, CA, USA) withdifferent settings for spores (SP), germinating spores (GSP) andvegetative cells (C2h and C12h). SP and GSP were processed fourtimes for 50 s each in the FastPrep machine at setting 6.5 m/s using0.1 mm zirconia/silica beads (Biospec Products Inc., Bartlesville,USA); C2h and C12h cells were bead-beated one time for 50 s at6.5 m/s. Samples were kept on ice during the intermediate pro-cessing steps. RNA samples were finally resuspended in 40 ml ofRNase-free water and rapidly frozen at�80 �C. RNA quantities weredetermined with the Qubit Quantitation Platform (Invitrogen,Paisley, UK), while the quality was assessed with RNA 6000 NanoLab-Chip (Agilent Technologies, Santa Clara, CA, USA) using a 2100Bioanalyser (Agilent Technologies, Santa Clara, CA, USA).

2.6. Relative quantification of enterotoxin gene expression in CPMmodel

The mRNA level changes of enterotoxin genes hblC, nheC, andcytK in the different tested stages were analyzed in real-time RT-PCR. For relative quantification, the 16S rRNA gene was analyzed ashousekeeping gene by amplification with the P1eP4 primers pair.The oligonucleotide primer sets used for reference and target genesamplification are shown in Table 1. Primer sets for enterotoxingenes were designed against the complete nucleotide sequence, asdeposited on GenBank, using LightCycler Probe Design Software 2.0(Roche, Indianapolis, IN, USA). The optimum annealing

temperature for each primer set was determined prior to theanalysis of experimental samples. Reverse transcription was per-formed using 200 ng DNA-free RNA, random primers, and theTranscriptor First Strand cDNA Synthesis kit (Roche, Indianapolis,IN, USA), following the supplier's recommendations. A samplevolume of 20 ml was used for all quantification assays, which con-tained a 1� final concentration of SYBR green PCR master mix,0.5 mM gene specific primers, and 1 or 2 ml of template. Sampleswere heated at 95 �C for 10 min before cycling for 45 cycles of 95 �Cfor 10 s, 56 �C or 57 �C for 20 s, and 72 �C for 25 s. In each step thetemperature transition rate was 20 �C/s. A melting curve plotted atthe end of each run verified the specificity of the amplificationproduct. For each independent experiment, three technical repli-cates were run in real-time PCR analysis. Prior to quantitative an-alyses, standard curves were constructed for housekeeping andtarget genes using cDNA from log B. thuringiensis UC10070 cellsgrown in BHI medium: serial decimal dilutions of the cDNA withnuclease-free water were used to generate standard curves fortarget genes, as well as with P1eP4 oligonucleotides, whichcovered 3e5 orders of magnitude in the range of the samples inorder to calculate the specific efficiency (E) using LightCyclerSoftware 3.5. Relative expression levels between samples werethen calculated as fold changes, where each PCR cycle represents atwo-fold change (Livak and Schmittgen, 2001).

2.7. RNA amplification and hybridization

The complete genome sequence of B. thuringiensis sv konkukianstr. 97e27 (NCBI Reference Sequence: NC_005957.1) was chosen forits high homology with B. thuringiensis UC10070, to design probescorresponding to 5197 genes spotted in duplicates onto Elec-traSenseH 12K microarrays chip (CombiMatrix Irvine, CA, USA).Probes were designed using the Combimatrix Automated Probe-Design Suite (CombiMatrix, Irvine, CA, USA). Thirteen negativecontrol probes designed on three Arabidopsis genes were synthe-sized in 30 replicates randomly distributed on the chip as previ-ously described (Bassi et al., 2013). Nucleotides quantity and qualitywere determined by NanoDrop® ND-1000 (ThermoFisher, Wal-tham, MA, USA) and analyzed on a RNA 6000 Nano LabChip (Agi-lent Technologies, Santa Clara, CA, USA) using a 2100 bioanalyzer(Agilent Technologies, Santa Clara, CA, USA).

RNA amplification, labelling, fragmentation and hybridizationwere performed using the MessageAmp™ II-Bacteria Kit for

D. Bassi et al. / Food Microbiology 55 (2016) 73e8576

prokaryotic RNA (Ambion, ThermoFisher, Waltham, MA, USA), alinear in vitro transcription based RNA amplification system (VanGelder et al., 1990) as previously described (Bassi et al., 2013).Slides previously coated with an imaging solution were scannedwith a ScanArray 4000 XL (PerkineElmer, USA) and raw data wereextracted with ScanArray Express 4.0 and Microarray Imager soft-ware (CombiMatrix, Irvine, CA, USA). Experiments were carried outin three biological replicates per stage. Each gene was present atleast in duplicate on the slide, so each biological replicate was hy-bridized in duplicate on the same microarray slide.

2.8. Microarray data analysis

The fluorescence signal for Cy5-dye channel and backgroundsubtractions were determined with Microarray Imager software(CombiMatrix, Irvine, CA, USA). The fluorescence signal of each spotwas calculated as the difference between the mean of pixel in-tensities and the mean of background fluorescence signals, definedby surrounding pixel intensity. Signals were log2 transformed anddifferentially expressed genes in the different conditions testedwere identified with a one-way ANOVA test, with P value < 0.05and for induction or repression ratio equal or higher than 1-fold. Apreliminary quality control of the data was also carried out withprincipal component analysis (PCA). Filtered data were thenanalyzed using Microarray Expression Viewer software (Saeedet al., 2006); in order to identify groups of genes with similartranscription profiles, the significantly regulated genes were sub-divided into clusters with different expression patterns by using K-means clustering (Soukas et al., 2000). Significance Analysis ofMicroarrays (SAM)with a one-class design (Tusher et al., 2001) wasapplied to calculate mRNA abundance in dormant spores.

For a functional interpretation of the transcriptional activity, theB. thuringiensis sequence sv. konkukian str. 97-27 genome annota-tions (http://img.jgi.doe.gov/) was used. Six groups of functionallyrelated genes were identified and ordered by HCL hierarchicalclustering using Microarray Expression Viewer software (Saeedet al., 2006).

2.9. Microarray data accession number

The data discussed in this publication have been deposited inNCBI's Gene Expression Omnibus (Edgar et al., 2002) and areaccessible through GEO Series accession number GSE65259 (http://www.ncbi.nlm.nih.gov/geo/query/acc.cgi?acc¼GSE65259).

3. Results and discussion

3.1. Characterization of the food isolate Bacillus strain UC10070

Bacillus strain UC10070 isolated from a biofilm on a spoiltvegetable based puree from organic farming was used for thisstudy. It was first of all characterized by means of 16S rRNA genesequencing and assigned accordingly to B. cereus sensu lato. Scan-ning electron microscopy (SEM) analysis revealed the presence oftoxin crystals (Fig. 1A) and PCR reaction confirmed the positivity ofthe strain to cry4 and cry4a1 genes, therefore indicating that thestrain belongs to B. thuringiensis. To clarify the presence and theexpression of genes coding for B. cereus toxins, PCR and reversetranscriptase PCR were performed on DNA and RNA extracted fromvegetative cells using primers pairs (Table 1) targeted to the hbl andnhe operons. These experiments demonstrated that UC10070 strainharbours and expresses the genes coding for the two multipartiteenterotoxins Hbl and Nhe and for cytotoxin K. The production of theL2 component of Hbl enterotoxin, involved in the diarrhoeal syn-drome, was also assessed in the supernatant of UC10070 broth

culture by immunological assay and gave positive results. Albeitsome limitations of the BCET-RLPA kit employed were recentlyoutlined (Kaminska et al., 2014), the immunological and PCR resultshere obtained are in accordance, and confirm the enterotoxinsproduction of the strain UC10070. The PCR approach, amplifyingthe ces gene, demonstrated the absence of the non-ribosomalpeptide synthetase gene responsible for the cereulide production.

3.2. Food model development to study B. thuringiensis UC10070germination kinetics and growth

A number of preliminary experiments were carried out in orderto: (i) identify a food substrate which supported the growth andgermination of B. thuringiensis UC10070; (ii) mimic the handlingconditions of a ready to eat food, (iii) follow the germination ki-netics in the food models and (iv) select the key steps to reproduceand study gene expression during germination, outgrowth and cellcycle of inoculated B. cereus sensu lato in a food model. Threecommercial vegetable soups based on pepper, artichoke andspinach, and one experimental pasteurized soup (CPM), weretested. A thermal treatment for 15 min at 80 �C, and incubation inanaerobic conditions, were used to reproduce, in our experimentalfoodmodels, the steps of an industrial process. Chemical analysis ofthe primary factors affecting microbial growth, pH and water ac-tivity (aw), allowed a first selection of the best model to use foranalysis. Pepper cream presented an optimum value of aw (0.97), asthe CMP model, but the pH (4.63) seemed to be too restrictive tosupport the growth of B. thuringiensis. The cream of artichoke andspinach showed pH values of respectively 5.63 and 5.94 suitable forthe development of the bacterium, but an aw of 0.94 that probablylimited spores germination and the growth of vegetative cells. Theanalysis of growth kinetics carried out in three replicates indicatedthat the CPM soup substrate (pH 5.95 and aw 0.97) better supportedthemultiplication of vegetative cells at both 20 �C (Fig. 2A) and 4 �C(Fig. 2B), the germination of spores and their subsequentoutgrowth: this substrate was therefore selected as the mostappropriate artificial foodmodel and the germination kinetics wereanalyzed in this medium.

The selection of the crucial steps for B. thuringiensis UC10070spores germination and outgrowth in CPM soup stored at roomtemperature was performed on the basis of optical microscopytogether with scanning electron microscopy (SEM) and EDAX mi-croanalyses results (Bassi et al., 2009). At room temperature, after20 min from thermal treatment, the optical microscopic analysis offood samples, based on the observation of at least 15 fields, showedthat the transition of Bacillus spores from phase-bright to phase-dark was commenced. This step coincides with the release of cal-cium dipicolinate from spores, the entrance of water into the coreenvironment and consequent increase in volume (Paidhungat et al.,2000; Setlow, 2003). After 40 min about the 90% of spores showedthe conversion to the dark germinating phase. This observationwere confirmed by SEM analysis which allowed the identificationof morphological changes of B. thuringiensis spores and cells.Dormant spores (Fig. 1 A and B) in CPM food model, afterpasteurization, underwent morphological changes in their overallstructure after 20 min (Fig. 1C) with an increase cell dimension(average spore length 1.4 ± 0.09 vs 1.2 ± 0.10) and a more elongatedstructure. To further investigate the germination process, X-raymicroanalysis (EDAX) (Hayat, 1980; Bassi et al., 2009) wasemployed together with the high resolution carbon-coating SEMtechnique. This approach allowed to identify calcium at the cellularlevel and to monitor calcium-dipicolinate release (Ca-DPA) fromspores, an event that is considered a key step in the germinationprocess (Santo and Doi, 1974; Vepachedu and Setlow, 2007). Therelease of calcium dipicolinate started already 20 min after

Fig. 1. Scanning electron microscopy (SEM) images using the gold coating technique; pictures show representative fields selected among all the performed observations: A, anenterotoxin crystal protein of B. thuringiensis UC10070; B, dormant spores of B. thuringiensis UC10070 in CPM food model at the beginning of the experiment (average length1.2 ± 0.10); C, spores after 20 min (average length 1.4 ± 0.09); D, germinating spores at 40 min after heat treatment; E, vegetative cells at 2 h; F, log phase cells at 12 h. Arrowsindicate: CP, crystal protein; DS, dormant spore; GS, germinating spore; VC, vegetative cell.

D. Bassi et al. / Food Microbiology 55 (2016) 73e85 77

pasteurization with approximately the 30% of the spores havinglost the Ca2þ content (Fig. 3). The germination process wascompleted (>90% of the spores) 40min after the thermal treatment,when the spores had completely lost Ca2þ and adopted an elon-gated shape (Fig. 1D). At 120 min only vegetative cells that initiatedthe duplication, as demonstrated by the septa formation, weredetected (Fig. 1E). At 12 h the log phase vegetative cells did notenter yet in the sporulation process (Fig. 1F).

These microscopic examinations allowed to identify the criticaltime-points that were used in the concomitant transcriptomic an-alyses of B. thuringiensis UC10070: the spore dormant state (S), theonset of germination 40 min after heat activation (GSP), thebeginning of the outgrowth process at 2 h after heat activation(C2h), and vegetative cells at 12 h after heat treatment (C12h).

3.3. Microarray analysis of B. thuringiensis UC10070 geneexpression during germination and growth cycle in an artificiallycontaminated vegetable soup model

A custom array analysis was performed on total RNA extractionsat the above-described conditions and time-points S, GSP, C2h andC12h to determine genes differentially expressed in B. thuringiensisUC10070 grown in CPM food model. Even though we used avegetable food model to perform our experiments, the monitoredtime-points for the germination assay were consistent with theones observed in B. subtilis in germination medium (Keijser et al.,2007) and in B. cereus with L-Alanine (Wang et al., 2011). RNA

from sporulating cells was not considered in the course of the studydue to the lack of reliable methods to differentiate mRNA of themother cell from transcripts in the prespore. For each sampling,cells were rapidly harvested and frozen in order to stabilize thenucleic acids in the different steps of the biological cycle. RNA wasisolated efficiently from dormant spores and more easily fromgerminating spores and vegetative cells adopting a combination ofa bead-beating mechanical method using zirconia beads, with acolumn RNA purification in order to preserve its integrity (Bassiet al., 2013). Gene expression was found to be highly dynamicduring the B. thuringiensis UC10070 life cycle and was found toinvolve a large number of genes, as demonstrated by the PrincipalComponent Analysis (data not shown). A total of 1646 genes werefound to be differentially expressed and modulated in the fourdistinct analyzed conditions (Fig. 5). The large amount of genesdifferentially expressed was consistent with the relevant metabolicand morphologic changes undergoing along the Bacillus life cycle.

3.4. Transcripts in spores

The general assumptions that spores do not contain RNA tran-scripts have been now rejected; all the transcriptional studiespublished in the last years showed that mRNA molecules are pre-sent in these dormant entities and the hypothesis of a functionalrole is proposed but not yet confirmed (Alsaker and Papoutsakis,2005; Fawcett et al., 2000; Liu et al., 2004; Segev et al., 2013;Setlow, 2003; Wang et al., 2006). In addition to the fact that

Fig. 2. Kinetics of growth of B. thuringiensis UC10070 observed in four different vegetables mix at room temperature of 20 �C (A) and refrigerator temperature of 4 �C (B). Numbersare expressed as the mean values of triplicate experiments, with error bars; A artichoke, - pepper, : spinach, C CPM model.

Fig. 3. SEM (left) and X-ray microanalysis (right) maps of the same picture at incubation time of 20 min. The spores A, B and C showed no calcium inside, as represented by theabsence of a fluorescent spot. These spores have already triggered the germination process.

D. Bassi et al. / Food Microbiology 55 (2016) 73e8578

spore mRNA composition is clearly dissimilar from that of vegeta-tive cells, differences have been found also in ribosomal RNA spe-cies. As observed in B. subtilis (Keijser et al., 2007), C. novyi(Bettegowda et al., 2006), C. sporogenes (Bassi et al., 2013) andC. difficile (Dembek et al., 2013), B. thuringiensis UC10070 presentedevidence of a truncated 23S rRNA in spores that cannot be detectedin the vegetative cell forms (Fig. 4). In fact, in germinating spores at

40 min this processed rRNA species has already disappeared, a signof an advanced germination phase evolving to outgrowth. Thespecific role of this shorter rRNA is not known, but its presence inall the analyzed species of Bacillus and Clostridium, and its disap-pearance after the first steps of germination suggests a possibleregulatory function associated with the spore cycle. According toSegev et al. (2012), rRNA degradation seems to be related to the

S1

S2

S3

GSP1

GSP2

GSP3

C2H1

C2H2

C2H3

C12H

1

C12H

2

C12H

3

16S 23S

5S

tr 23S

Fig. 4. RNA integrity measured with Agilent 2100 Bioanalyser (Agilent Technologies).Two major bands (16S and 23S rRNA subunits) were present in all samples in the fouranalyzed stages. An extra band corresponding to a truncated 23S rRNA (tr 23S)appeared only in all the three replicates of dormant spores (samples 1, 2, 3), while wasabsent in GSP at 40 min after heat treatment (samples 4, 5, 6), after 2 h (sample 7, 8, 9)and after 12 h (samples 10, 11, 12).

D. Bassi et al. / Food Microbiology 55 (2016) 73e85 79

dormant state to minimize normal activity and enter dormancy,and the small rRNA fragments could be a result of this growth ar-rest. All this process can be considered an adaptation to a preciseenvironment where the spore finds itself and has to prepare all the“ready-to-use kit” of molecular tools to germinate (Segev et al.,2012).

In contrast to other reported studies on spore transcripts, inB. thuringiensis UC10070 a high number of transcripts (824) wasfound in the dormant spores used to inoculate the food model andcollected immediately after. This was a relatively abundant numberif compared to what observed in previous studies in B. cereus and inB. subtilis, where respectively 46 (Van Melis et al., 2011), 23 (Keijseret al., 2007) and 369 transcripts (Segev et al., 2012) were detected.Segev et al. (2012) studied the rRNA dynamics in ageing spores andobserved that in B. subtilis, the number of mRNAs in spore isinversely related to the spore age and that the RNA degradation isaccelerated in spores held at high temperatures.

In our experiments, the spores were prepared, cooled to 4 �Cand directly used for transcription studies. This might explain thehigher number of transcripts detected in UC10070 and its rapidentrance in germination up to favourable environmental condi-tions. The mRNAs found in the spore transcriptome coded for genesbelonging to DNA metabolism, transport, transcription, translationand other metabolic pathways (sporulation, cell wall metabolism,energy production), consistently with what already described inB. subtilis and its relatives, C. novyi and C. sporogenes (Bassi et al.,2013; Bettegowda et al., 2006; Keijser et al., 2007).

Among the total 824 transcripts, we analyzed the most abun-dant genes identified in spores of B. thuringiensis UC10070. The sspImRNA for small-acid-soluble protein A, a protein responsible forDNA protection from chemical and physical insults during thepassage to spore, and the transcript for YhcN/YlaJ, a foresporespecific sporulation lipoproteinwere found at this stage, being bothprobably a rest of the gene expression during the sporulation stage.Another transcript present in spores was a thioredoxin familyprotein; since in B. subtilis thiol-disulfide oxidoreductases seemedto have a role in cortex synthesis and consequent spore maturation(Erlendsson et al., 2004), we speculate also in this case an entrap-ping of transcripts during sporulation phases. Interestingly, othergenes whose transcripts were quite abundant in spores included apossible PAP2 BcrC_like subfamily protein, a bacitracin transportpermease involved in bacitracin resistance and a gene coding for aMajor Facilitator Superfamily transporter; the first is an ABCtransporter which pumps out bacitracin from the cell wall (Cao andHelmann, 2002) and the second a drug anti-porter that pumps a

variety of solutes, including drugs, across the membrane (Han et al.,2006). The prevalence in spores of these drug transporter tran-scripts may be correlated to previous spore maturation or either tothe transport of metabolites necessary for the germination process.

AnmRNA encoding for a transposase IS660was also abundant inspores. Insertion sequences may play an important role in internalgenetic rearrangements in the genome, likely to support themechanisms of adaptation to extreme environments, such as thosewith extreme values of pH, temperature, pressure, or salinity(Takami et al., 2001). In this case, its functionwas difficult to explainin spores since it could be related to a former insertional event.

A 50% of transcripts found in spores coded for hypotheticalproteins or proteins with unknown function as seen in previousstudies (Bettegowda et al., 2006; Dembek et al., 2013) and arisingthe need to deeply investigate on their role in cell cycle of sporeformers.

3.5. Functional analysis and hierarchical clustering duringgermination

The presence of 407 (120 in cluster 2 and 287 in cluster 3) up-regulated genes in GSP vs SP reflected the transfer of thedormant spore to an active state (Fig. 5). This initiates a cascade ofprocesses that gradually degrade the protective structures of thespore and resume cellular processes and its metabolism, ultimatelyleading to the vegetative cell. In accordance with previously pub-lished data on the temporal gene expression during Bacillus spp.spores outgrowth (Keijser et al., 2007), important house-keepinggenes, encoding proteins such as translation initiation factors, ri-bosomal proteins, and elongation factors, were found to be tran-scribed in the first 40 min after heat activation of dormant spores,together with ample modulation of their expression levels duringoutgrowth and vegetative cycle.

In order to have a functional interpretation of the transcriptionalactivity during the germination process and outgrowth, groups offunctionally related genes were identified using B. thuringiensis 97-27 genome annotations and ordered by hierarchical clustering.Functional categories were then analyzed in relation to theirexpression patterns (Fig. 6) and the most representative genes foreach categories have been evaluated (Fig. 7).

3.5.1. TransportOne of the major significantly over-represented group of func-

tional genes identified encoded for proteins involved in transport ofvarious molecules (116 differently regulated genes). ABC trans-porters specific for ions, sugars, and other organic compounds likedrug/metabolites, are considered essential to restart the normalvegetative behaviour after dormancy (Dembek et al., 2013); thesetranscripts were mostly found to be present in spores and theywere differentially regulated during germination (Fig. 7). Amongthese, 32 were up-regulated at 40 min and, particularly, the onescoding for spermidine/putrescine ABC transporters, ferrous irontransport proteins, zinc transporters, cobalt transport proteins,glycerol-3-phosphate ABC transporters. The involvement of ionsduring the germination process has been studied since the very firstworks by Levinson and Hyatt (1963). It could be explained as asupport to germinants action reinforcing their effect; moreover, ithas been supposed that the suspension of the ion channels action iscritical for spore germination (Mitchell et al., 1986). The importanceof metal ion homeostasis and particularly of iron was evidenced bythe fact that eight genes for iron-regulated transport proteins werefound to be significantly expressed during germination.

Transcripts for glycine-betaine transport protein OpuAB, awidely diffuse bacterial system for cell protection against high so-lute concentration, were found in spores and remained present in

Fig. 5. Cluster graphics of the 4 studied stages. Using K-means clustering, the filtered genes were grouped into 5 different clusters of expression; the bars indicate the centroid value,which is the mean value in every group (in order: S, GSP, 2 h, 12 h). The Y-axis shows the normalized intensities (�1, 0, 1).

D. Bassi et al. / Food Microbiology 55 (2016) 73e8580

GSP, suggesting that osmotic defence is important in the earlieststages of outgrowth, in agreement with previous studies ongermination process (Keijser et al., 2007). A large number oftransporter genes encoding putative multidrug transporters, suchas SMR and Bcr/CflA family, which were up-regulated during thefirst 40 min of outgrowth, may also provide the germinated sporewith a transient resistance against antimicrobial compounds.

3.5.2. Transcriptional regulation and protein synthesisMany genes involved in transcriptional regulation for rapid re-

covery of cell functionswere found to be already present among thespore transcript pool. The end of dormancy is a process that needsthe progressive re-starting of transcription and translation pro-cesses. During the first 40 min, several ribosomal proteins were up-regulated: among these, we found twenty-seven 50S ribosomalproteins and nineteen 30S subunit proteins (Fig. 7). This evidencecan be interpreted on the basis of the data obtained by Segev et al.(2012); according to this work, to achieve quiescence, spores ofB. subtilis degrade their RNA to reduce their cellular activity. At theonset of germination, transcription could restart within the sporeand rRNA resynthesized upon germination maintaining the samehigh level also during vegetative life. In addition, genes for RNApolymerase subunits a, b and d were up-regulated during germi-nation together with genes for transcriptional regulators. A largenumber of genes coding for proteins involved in the control ofdifferent biological processes were found. Particularly, 22 genesencoding for transcriptional regulators were up-regulated duringgermination; examples were members of TetR, MarR, MerR(especially found in C12h) transcription factors families (TFs), that

were identified as common families to Bacillales, Lactobacillales andClostridiales (Moreno-Campuzano et al., 2006) in addition to GntR,ArsR, DeoR, Fur families of transcriptional regulators. These pro-teins are responsible for rapid adaptation of bacteria to changingenvironmental conditions, including resistance to different anti-microbial compounds and oxidative stress agents, controlling theexpression of drug efflux pumps. Members of GntR family proteins,also found in dormant spores, respond to environmental changesaffecting the carbohydrate metabolism of the cell and may providebacteria with the ability to grow in the presence of several carbonsources and to rapidly adapt their gene expression to changingnutrient conditions (Reizer et al., 2006). Transcripts coding for thefumarate and nitrate reduction regulator CRP-Fnr, member of thecyclic AMP receptor proteins, were present both in S and GSP; theyare known to play an important role in modulating the expressionof many metabolic genes in several facultative or strictly anaerobicbacteria; their functions also include the control of virulence factors(Baltes et al., 2005; Bartolini et al., 2006). In addition, transcripts forRNA polymerase s-factor, ECF type, that controls genes involved incell envelope functions (protein transport and secretion processes),in response to extra-cytoplasmic stress in B. subtilis and Bacilluslicheniformis (Wecke et al., 2006), and RNA polymerase sK factor,responsible for the expression of sporulation specific genes in themother cell, were found in dormant spores.

During outgrowth and cell vegetative growth (C2h and C12h),genes involved in DNA replication, transcription (DNA polymeraseIII, DNA-directed RNA polymerase delta subunit) and RNA trans-lation, like transcription elongation antitermination factor proteinsNusB and NusG, whose function was described in B. subtilis

Fig. 6. Gene expression profiles in the four steps of B. thuringiensis UC10070 life cycle in different functional categories. A) All genes; B) Transport; C) Ribosomal proteins; D)Transcription; E) Germination; F) Membrane biosynthesis and motility; G) DNA repair.

D. Bassi et al. / Food Microbiology 55 (2016) 73e85 81

(Yakhnin et al., 2008), were over-represented. During the latestages of outgrowth, cells appeared to prepare for septation, asindicated by the overexpression of septation ring formation regu-lator ezrA (Errington, 2001), giving evidence that cells are in anactive growing phase. Transcripts for the transcriptional regulatorPlcR (Phospholipase C Regulator) were found to be up-regulated12 h after germination. This transcriptional regulator takes part inthe control of most known virulence factors in B. cereus (entero-toxin, haemolysins, phospholipases and proteases) (Gohar et al.,2008), acting as a quorum-sensing system whose molecularmechanism has been explained by Grenha et al. (2013). To beactivated, PlcR needs its cognate PapR peptide, whose transcriptwas also found to be overrepresented in vegetative cells. PlcR is thefirst example described of a pleiotropic regulator required for thecontrol of extracellular virulence factor expression in pathogenicBacillus spp. (Agaisse et al., 1999) and seemed to be present in all

the members of the B. cereus sensu lato, except for B. anthraciswhere the PlcR regulator is present but inactive (due to a non-sensemutation).

3.5.3. Cell wall biosynthesis and motilityExpression pattern analysis of genes coding for proteins

involved in cell wall biosynthesis, revealed in dormant spores thepresence of mRNAs for transpeptidase enzymes (D-alanyl-D-alaninecarboxypeptidase family proteins) that cross-link the peptido-glycan chains to form rigid cell walls. During germination, the newemerging cell has to synthesize new cell wall: accordingly, D-Ala D-Ala carrier protein ligase, UDP-N-acetylmuramate-L-alanine ligase,cell wall hydrolase possible N-acetylmuramoyl-L-alanine amidase,a glycosyltransferase, murein hydrolase export regulator, were up-regulated at 40 min in B. thuringiensis UC10070 (Fig. 7).

Motility is known to be absent in dormant spores, while is a

Fig. 7. Functional categories and expression patterns of the most representative genes modulated during germination and outgrowth phases.

D. Bassi et al. / Food Microbiology 55 (2016) 73e8582

prerequisite in the vegetative life; in B. thuringiensis UC10070 asingle transcript for the flagellar motor switch protein was presentin the spore phase, while the majority of transcripts for flagellarbiosynthesis-assembly proteins and chemotaxis were overex-pressed at C2h and C12h.

3.5.4. DNA repairA number of general DNA repair genes were expressed actively

during vegetative life (C12h). At the same time some transcriptscoding for proteins involved in more specific repair activity couldbe found in dormant spores. DNAmismatch repair protein and DNArepair protein RadC, that was supposed to be involved in the pro-tection of spores against harsh environmental conditions anddamaging following X- and UV-irradiation (Felzenszwalb et al.,1986) were an example (Fig. 7). DNA of dormant spores isbelieved to be in a supercoiled state, providing protection againstdamage. The helicase activity of Holliday junction RuvB, andmethionine gamma-lyase, observed during early stages of germi-nation, might be necessary to relax rapidly the supercoiled DNA

allowing an efficient reactivation of transcription, likely after thepartial degradation of the SASPs that coat and protect DNA duringspore dormant state by complete saturation and tight binding,changing the structure to a more stable conformation (Setlow,1995).

3.6. Expression of enterotoxin genes in the food model

Although limited evidence on B. cereus toxins expression in foodis available (Delbrassinne et al., 2011; Rajkovic et al., 2013), en-terotoxins production has been demonstrated to be highly dynamicdepending on pH, glucose availability and oxygen. Previous studiesreported that the highest toxin levels can be achieved during thelate log-early stationary phase for cultures grown in broth and thatno significant increase in toxin production occurs during the latestationary phase (Mckillip, 2000); more recently, Madeira et al.(2015), studying the exoproteome of B. cereus found that the ma-jority of toxin-related proteins were produced during the expo-nential growth phase. In fact, when B. cereus reaches high cellular

Table 2Relative quantification of the expression of virulence genes by means of qRT-PCR.Data (average ± standard deviation) are expressed as fold changes in mRNA levels.GSP, germinating spores; C2h, vegetative cells two hours after the heat treatment;C12h, vegetative cells at mid-log phase.

GSP C2h C12h F value

hblC 0.69 ± 0.18 b 2.70 ± 0.62 a 1.65 ± 0.24 b 19.251**nheC 1.05 ± 0.24 b 2.17 ± 0.25 a 2.01 ± 0.46 a 9.48*cytK 0.69 ± 0.39 a 0.93 ± 0.37 a 1.42 ± 0.48 a 2.30 ns

* <0.05, ** < 0.01, *** <0.0001, ns ¼ not significant.

D. Bassi et al. / Food Microbiology 55 (2016) 73e85 83

density, different toxic compounds are produced: the two enter-otoxic complexes, haemolysin BL (HBL) and non-haemolyticenterotoxin (NHE), several phospholipases-C, a collagenase andseveral haemolysins/cytolysins (HlyI, HlyII, HlyIII and CytK)(Ramarao and Sanchis, 2013). Coherently with these observations,the transcriptome analysis of B. thuringiensis UC10070 demon-strated no signal during germination for probes related to genescoding for Hbl and Nhe enterotoxic complexes, Haemolysin II,CryIA, CytK and haemolysin type III. In contrast, at 12 h after heatactivation, genes coding for delta-endotoxin CryIA, cytotoxin K,haemolysin type III, and possible phospholipases, were signifi-cantly expressed. The analysis of microarray data showed that theexpression of both transcripts of nhe and hble operons weresignificantly increased (P < 0.05) from GSP to C2h, but the foldincrease was lower than the defined threshold. To better study theenterotoxin expression, the total RNA from the same samples wasanalyzed by quantitative RT-PCR, evaluating hblC, nheC and cytKmRNA levels, during the B. thuringiensis UC10070 life cycle(Table 2). The transcript of hbl and nhe operons were found to in-crease in the passage from GSP to C2h, with values respectivelythreefold and twofold greater than in germinating spore (GSP). Theexpressions were maintained in the log phase (C12h) cells. Simi-larly transcripts corresponding to cytK gene, raised more graduallyreaching the maximum expression at C12h interval, a stage inwhichmost of the genes involved in themetabolic activity are over-represented in themicroarray analysis. To confirm the trend for Hblpresence, we further assessed in CPM model at the stages GSP, 2 hand 12 h the production of the L2 component of HBL enterotoxin byimmunoassay. The toxin was detected in low concentration twohours after spore activation and its amount increased over timereaching the higher value at 12 h. This is in agreement with studieson Hbl characterization, where a growth period of 5e6 h was usedfor in vitro routine production of enterotoxins (Beecher and Wong,1994; Dietrich et al., 1999). These data indicate that B. thuringiensisproduces enterotoxins in a vegetable food model, with a kineticsimilar to that described in B. cereus (Garcia-Arribas and Kramer,1990; McKillip, 2000).

4. Conclusions

Transcriptomics was demonstrated to be not only a powerfultool to study the germination and outgrowth of Bacillus spores, butalso a suitable method to assess the environmental response ofbacterial pathogens in foodstuffs. Employing for the first time avegetableebased foodmodel, we showed that under the conditionsof high water activity and availability of nutrients, B. thuringiensisgerminates, grows and expresses enterotoxin genes. This workprovides new basic knowledge on B. cereus sensu lato and extendsour knowledge on the metabolic versatility of these bacteria.

Acknowledgements

The research was supported by grants from the Regione

Lombardia founding scheme “GENOBACT” project n.G41J10000400002.

References

Abee, T., Groot, M.N., Tempelaars, M., Zwietering, M., Moezelaar, R., Van DerVoort, M., 2011. Germination and outgrowth of spores of Bacillus cereus groupmembers: diversity and role of germinant receptors. Food Microbiol. 28,199e208.

Agaisse, H., Gominet, M., Økstad, O.A., Kolstø, A.B., Lereclus, D., 1999. PlcR is apleiotropic regulator of extracellular virulence factor gene expression in Bacillusthuringiensis. Mol. Microbiol. 32, 1043e1053.

Alsaker, K.V., Papoutsakis, E.T., 2005. Transcriptional program of early sporulationand stationary-phase events in Clostridium acetobutylicum. J. Bacteriol. 187,7103e7118.

Baltes, N., Mohamed, N., Jacobsen, I.D., Maas, A., Buettner, F.F.R., Gerlach, G.F., 2005.Deletion of the anaerobic regulator HlyX causes reduced colonization andpersistence of Actinobacillus pleuropneumoniae in the porcine respiratory tract.Infect. Immun. 73, 4614e4619.

Bargabus, R., Zidack, N., Sherwood, J., Jacobsen, B., 2002. Characterisation of sys-temic resistance in sugar beet elicited by a non-pathogenic, phyllosphere-colonizing Bacillus mycoides, biological control agent. Physiol. Mol. Plant P 61,289e298.

Bartolini, E., Frigimelica, E., Giovinazzi, S., Galli, G., Shaik, Y., Genco, C., Welsch, J.A.,Granoff, D.M., Grandi, G., Grifantini, R., 2006. Role of FNR and FNR-regulated,sugar fermentation genes in Neisseria meningitidis infection. Mol. Microbiol. 60,963e972.

Bartoszewicz, M., Kroten, M.A., Swiecicka, I., 2013. Germination and proliferation ofemetic Bacillus cereus sensu lato strains in milk. Folia Microbiol. 58, 529e535.

Bassi, D., Cappa, F., Cocconcelli, P.S., 2009. A combination of a SEM technique and X-ray microanalysis for studying the spore germination process of Clostridiumtyrobutyricum. Res. Microbiol. 160, 322e329.

Bassi, D., Cappa, F., Cocconcelli, P.S., 2013. Array-based transcriptional analysis ofClostridium sporogenes UC9000 during germination, cell outgrowth and vege-tative life. Food Microbiol. 33, 11e23.

Beecher, D.J., Wong, A.C., 1994. Improved purification and characterization of he-molysin BL, a hemolytic dermonecrotic vascular permeability factor from Ba-cillus cereus. Infect. Immun. 62, 980e986.

Berka, R.M., Hahn, J., Albano, M., Draskovic, I., Persuh, M., Cui, X., Sloma, A.,Widner, W., Dubnau, D., 2002. Microarray analysis of the Bacillus subtilis K-state: genome-wide expression changes dependent on ComK. Mol. Microbiol.43, 1331e1345.

Bettegowda, C., Huang, X., Lin, J., Cheong, I., Kohli, M., Szabo, S.A., Zhang, X.,Diaz, L.A., Velculescu, V.E., Parmigiani, G., 2006. The genome and tran-scriptomes of the anti-tumor agent Clostridium novyi-NT. Nat. Biotechnol. 24,1573e1580.

Cao, M., Helmann, J.D., 2002. Regulation of the Bacillus subtilis bcrC bacitracinresistance gene by two extracytoplasmic function sigma factors. J. Bacteriol. 184,6123e6129.

Carozzi, N.B., Kramer, V.C., Warren, G.W., Evola, S., Koziel, M.G., 1991. Prediction ofinsecticidal activity of Bacillus thuringiensis strains by polymerase chain reac-tion product profiles. Appl. Environ. Microbiol. 57, 3057e3061.

Ceragioli, M., Mols, M., Moezelaar, R., Ghelardi, E., Senesi, S., Abee, T., 2010.Comparative transcriptomic and phenotypic analysis of the responses of Ba-cillus cereus to various disinfectant treatments. Appl. Environ. Microbiol. 76,3352e3360.

Cowan, A.E., Koppel, D.E., Setlow, B., Setlow, P., 2003. A soluble protein is immobilein dormant spores of Bacillus subtilis but is mobile in germinated spores: im-plications for spore dormancy. P. Nat. Acad. Sci. 100, 4209e4214.

Daelman, J., Vermeulen, A., Willemyns, T., Ongenaert, R., Jacxsens, L.,Uyttendaele, M., Devlieghere, F., 2013. Growth/no growth models for heat-treated psychrotrophic Bacillus cereus spores under cold storage. Int. J. FoodMicrobiol. 161, 7e15.

Daffonchio, D., Cherif, A., Brusetti, L., Rizzi, A., Mora, D., Boudabous, A., Borin, S.,2003. Nature of polymorphisms in 16S-23S rRNA gene intergenic transcribedspacer fingerprinting of Bacillus and related genera. Appl. Environ. Microbiol.69, 5128e5137.

Delbrassinne, L., Andjelkovic, M., Rajkovic, A., Bottledoorn, N., Mahillon, J., VanLoco, J., 2011. Follow-up of the Bacillus cereus emetic toxin production in pennepasta under household conditions using liquid chromatography coupled withmass spectrometry. Food Microbiol. 28, 1105e1109.

Dembek, M., Stabler, R.A., Witney, A.A., Wren, B.W., Fairweather, N.F., 2013. Tran-scriptional analysis of temporal gene expression in germinating Clostridiumdifficile 630 endospores. PloS One 8, e64011.

Den Besten, H.M., Mols, M., Moezelaar, R., Zwietering, M.H., Abee, T., 2009.Phenotypic and transcriptomic analyses of mildly and severely salt-stressedBacillus cereus ATCC 14579 cells. Appl. Environ. Microbiol. 75, 4111e4119.

Dietrich, R., Fella, C., Strich, S., M€artlbauer, E., 1999. Production and characterizationof monoclonal antibodies against the hemolysin BL enterotoxin complex pro-duced by Bacillus cereus. Appl. Environ. Microbiol. 65, 4470e4474.

Dr�ean, P., McAuley, C.M., Moore, S.C., Fegan, N., Fox, E.M., 2015. Characterization ofthe spore-forming Bacillus cereus sensu lato group and Clostridium perfringensbacteria isolated from the Australian dairy farm environment. BMC Microbiol.

D. Bassi et al. / Food Microbiology 55 (2016) 73e8584

19, 15e38.Driks, A., 2002. Maximum shields: the assembly and function of the bacterial spore

coat. Trends Microbiol. 10, 251e254.Dworkin, J., Shah, I.M., 2010. Exit from dormancy in microbial organisms. Nat. Rev.

Microbiol. 8, 890e896.Edgar, R., Domrachev, M., Lash, A.E., 2002. Gene Expression Omnibus: NCBI gene

expression and hybridization array data repository. Nucleic Acids Res. 30,207e210.

EFSA, 2015. The European Union summary report on trends and sources of zoo-noses, zoonotic agents and food-borne outbreaks in 2013. EFSA J. 13.

Ehling-Schulz, M., Vukov, N., Schulz, A., Shaheen, R., Andersson, M., M€artlbauer, E.,Scherer, S., 2005. Identification and partial characterization of the non-ribosomal peptide synthetase gene responsible for cereulide production inemetic Bacillus cereus. Appl. Environ. Microbiol. 71, 105e113.

Ehling-Schulz, M., Fricker, M., Grallert, H., Rieck, P., Wagner, M., Scherer, S., 2006.Cereulide synthetase gene cluster from emetic Bacillus cereus: structure andlocation on a mega virulence plasmid related to Bacillus anthracis toxin plasmidpXO1. BMC Microbiol. 6, 20.

Erlendsson, L.S., M€oller, M., Hederstedt, L., 2004. Bacillus subtilis StoA is a thiol-disulfide oxidoreductase important for spore cortex synthesis. J. Bacteriol.186, 6230e6238.

Errington, J., 2001. Septation and chromosome segregation during sporulation inBacillus subtilis. Curr. Opin. Microbiol. 4, 660e666.

Fawcett, P., Eichenberger, P., Losick, R., Youngman, P., 2000. The transcriptionalprofile of early to middle sporulation in Bacillus subtilis. Proc. Nat. Acad. Sci. 97,8063e8068.

Felzenszwalb, I., Sargentini, N.J., Smith, K.C., 1986. Escherichia coli radC is deficient inthe recA-dependent repair of X-ray-induced DNA strand breaks. Radiat. Res.106, 166e170.

Garcia-Arribas, M.L., Kramer, J.M., 1990. The effect of glucose, starch, and pH ongrowth, enterotoxin and haemolysin production by strains of Bacillus cereusassociated with food poisoning and non-gastrointestinal infection. Int. J. FoodMicrobiol. 11, 21e33.

Gohar, M., Faegri, K., Perchat, S., Ravnum, S., Økstad, Gominet, M., Kolstø, A.,Lereclus, D., 2008. The PlcR virulence regulon of Bacillus cereus. PLoS One 3,e2793.

Grenha, R., Slamti, L., Nicaise, M., Refes, Y., Lereclus, D., Nessler, S., 2013. Structuralbasis for the activation mechanism of the PlcR virulence regulator by thequorum-sensing signal peptide PapR. Proc. Natl. Acad. Sci. U S A 110,1047e1052.

Guinebreti�ere, M.-H., Auger, S., Galleron, N., Contzen, M., De Sarrau, B., DeBuyser, M.-L., Lamberet, G., Fagerlund, A., Granum, P.E., Lereclus, D., De Vos, P.,Nguyen-The, C., Sorokin, A., 2013. Bacillus cytotoxicus sp. nov. is a novel ther-motolerant species of the Bacillus cereus group occasionally associated withfood poisoning. Int. J. Syst. Evol. Microbiol. 63, 31e40.

Han, C.S., Xie, G., Challacombe, J.F., Altherr, M.R., Bhotika, S.S., Bruce, D.,Campbell, C.S., Campbell, M.L., Chen, J., Chertkov, O., 2006. Pathogenomicsequence analysis of Bacillus cereus and Bacillus thuringiensis isolates closelyrelated to Bacillus anthracis. J. Bacteriol. 188, 3382e3390.

Hansen, B.M., Hendriksen, N.B., 2001. Detection of enterotoxic Bacillus cereus andBacillus thuringiensis strains by PCR analysis. Appl. Environ. Microbiol. 67,185e189.

Hayat, M.A., 1980. X-ray Microanalysis in Biology. University Park Press.Helgason, E., Økstad, O.A., Caugant, D.A., Johansen, H.A., Fouet, A., Mock, M.,

Hegna, I., Kolstø, A.-B., 2000. Bacillus anthracis, Bacillus cereus, and Bacillusthuringiensisdone species on the basis of genetic evidence. Appl. Environ.Microbiol. 66, 2627e2630.

Hendriksen, N.B., Hansen, B.M., 2006. Detection of Bacillus thuringiensis kurstakiHD1 on cabbage for human consumption. FEMS Microbiol. Lett. 257, 106e111.

Hoton, F.M., Andrup, L., Swiecicka, I., Mahillon, J., 2005. The cereulide genetic de-terminants of emetic Bacillus cereus are plasmid-borne. Microbiology 151,2121e2124.

Jensen, G.B., Hansen, B.M., Eilenberg, J., Mahillon, J., 2003. The hidden lifestyles ofBacillus cereus and relatives. Environ. Microbiol. 5, 631e640.

Jim�enez, G., Urdiain, M., Cifuentes, A., L�opez-L�opez, A., Blanch, A.R., Tamames, J.,K€ampfer, P., Kolstø, A.-B., Ram�on, D., Martìnez, J.F., Condo~ner, F.M., Rossell�o-M�ora, R., 2013. Description of Bacillus toyonensis sp. nov, a novel species of theBacillus cereus group, and pairwise comparisons of the group by means of ANIcalculations. Syst. Appl. Microbiol. 36, 383e391.

Kaminska, P.A., Yernazarova, A., Murawska, E., Swiecicki, J., Fiedoruk, K., Bideshi,D.K., Swiecicka, I. Comparative analysis of quantitative reverse transcriptionreal-time PCR and commercial enzyme immunoassays for detection of en-terotoxigenic Bacillus thuringiensis isolates. FEMS Microbiol. Lett. 357, 34e39.

Keijser, B.J., Ter Beek, A., Rauwerda, H., Schuren, F., Montijn, R., Van Der Spek, H.,Brul, S., 2007. Analysis of temporal gene expression during Bacillus subtilis sporegermination and outgrowth. J. Bacteriol. 189, 3624e3634.

Klijn, N., Weerkamp, A.H., de Vos, W.M., 1991. Identification of mesophilic lactic acidbacteria by using polymerase chain reaction-amplified variable regions of 16SrRNA and specific DNA probes. Appl. Environ. Microbiol. 1991 (11), 3390e3393.

Levinson, H.S., Hyatt, M.T., 1963. Bacterial spore germination with special referenceto Bacillus megaterium. Ann. N. Y. Acad. Sci. 102, 773e788.

Liu, H., Bergman, N.H., Thomason, B., Shallom, S., Hazen, A., Crossno, J., Rasko, D.A.,Ravel, J., Read, T.D., Peterson, S.N., 2004. Formation and composition of theBacillus anthracis endospore. J. Bacteriol. 186, 164e178.

Livak, K.J., Schmittgen, T.D., 2001. Analysis of relative gene expression data using

real-time quantitative PCR and the 2�DDCT method. Methods 25, 402e408.Madeira, J.P., Alpha-Bazin, B., Armengaud, J., Duport, C., 2015. Time dynamics of the

Bacillus cereus exoproteome are shaped by cellular oxidation. Front. Microbiol.22 (6), 342.

Markland, S.M., Farkas, D.F., Kniel, K.E., Hoover, D.G., 2013. Pathogenic psychroto-lerant sporeformers: an emerging challenge for low-temperature storage ofminimally processed foods. Foodborne Pathog. Dis. 10 (5), 413e419.

Mckillip, J.L., 2000. Prevalence and expression of enterotoxins in Bacillus cereus andother Bacillus spp., a literature review. A. Van Leeuw. J. Microbiol. 77, 393e399.

Mitchell, C., Iyer, S., Skomurski, J.F., Vary, J.C., 1986. Red pigment in Bacillus mega-terium spores. Appl. Environ. Microbiol. 52, 64e67.

Moir, A., Smith, D.A., 1990. The genetics of bacterial spore germination. Annu. Rev.Microbiol. 44, 531e553.

Moir, A., Corfe, B.M., Behravan, J., 2002. Spore germination. Cell Mol. Life Sci. 59,403e409.

Moreno-Campuzano, S., Janga, S.C., P�erez-Rueda, E., 2006. Identification and anal-ysis of DNA-binding transcription factors in Bacillus subtilis and other Firmi-cutes e a genomic approach. BMC Genomics 7, 147.

Nakashio, S., Gerhardt, P., 1985. Protoplast dehydration correlated with heat resis-tance of bacterial spores. J. Bacteriol. 162, 571e578.

Nicholson, W., Setlow, P., 1990. Sporulation, germination and outgrowth. In:Harwood, C.R., Cutting, S.M. (Eds.), Molecular Biological Methods for Bacillus.Wiley, New York, pp. 391e540.

Paidhungat, M., Setlow, B., Driks, A., Setlow, P., 2000. Characterization of spores ofBacillus subtilis which lack dipicolinic acid. J. Bacteriol. 182, 5505e5512.

Passalacqua, K.D., Varadarajan, A., Byrd, B., Bergman, N.H., 2009. Comparativetranscriptional profiling of Bacillus cereus sensu lato strains during growth inCO2-bicarbonate and aerobic atmospheres. PLoS One 4, e4904.

Priest, F.G., Barker, M., Baillie, L.W.J., Holmes, E.C., Maiden, M.C.J., 2004. Populationstructure and evolution of the Bacillus cereus group. J. Bacteriol. 186,7959e7970.

Rajkovic, A., Kljajic, M., Smigic, N., Devlieghere, F., Uyttendaele, M., 2013. Toxinproducing Bacillus cereus persist in ready-to-reheat spaghetti Bolognese mainlyin vegetative state. Int. J. Food Microbiol. 167, 236e243.

Ramarao, N., Sanchis, V., 2013. The pore-forming haemolysins of Bacillus cereus: areview. Toxins 5, 1119e1139.

Rasko, D.A., Altherr, M.R., Han, C.S., Ravel, J., 2005. Genomics of the Bacillus cereusgroup of organisms. FEMS Microbiol. Rev. 29, 303e329.

Reizer, A., Deutscher, J., Saier, M., Reizer, J., 2006. Analysis of the gluconate (gnt)operon of Bacillus subtilis. Mol. Microbiol. 5, 1081e1089.

Saeed, A.I., Bhagabati, N.K., Braisted, J.C., Liang, W., Sharov, V., Howe, E.A., Li, J.,Thiagarajan, M., White, J.A., Quackenbush, J., 2006. TM4 microarray softwaresuite. Method Enzymol. 411, 134e193.

Samapundo, S., Heyndrickx, M., Xhaferi, R., De Baenst, I., Devlieghere, F., 2014. Thecombined effect of pasteurization intensity, water activity, pH and incubationtemperature on the survival and outgrowth of spores of Bacillus cereus andBacillus pumilus in artificial media and food products. Int. J. Food Microbiol. 181,10e18.

Santo, L.Y., Doi, R.H., 1974. Ultrastructural analysis during germination andoutgrowth of Bacillus subtilis spores. J. Bacteriol. 120, 475e481.

Schmidt, T.R., Scott, E.J., Dyer, D.W., 2011. Whole-genome phylogenies of the familyBacillaceae and expansion of the sigma factor gene family in the Bacillus cereusspecies-group. BMC Genomics 12, 430.

Schnepf, E., Crickmore, N., Van Rie, J., Lereclus, D., Baum, J., Feitelson, J., Zeigler, D.R.,Dean, D.H., 1998. Bacillus thuringiensis and its pesticidal crystal proteins.Microbiol. Mol. Biol. Rev. 62, 775e806.

Segev, E., Smith, Y., Ben-Yehuda, S., 2012. RNA dynamics in aging bacterial spores.Cell 148, 139e149.

Segev, E., Rosenberg, A., Mamou, G., Sinai, L., Ben-Yehuda, S., 2013. Molecular ki-netics of reviving bacterial spores. J. Bacteriol. 195, 1875e1882.

Sergeev, N., Distler, M., Vargas, M., Chizhikov, V., Herold, K.E., Rasooly, A., 2006.Microarray analysis of Bacillus cereus group virulence factors. J. Microbiol. Met.65, 488e502.

Setlow, P., 1995. Mechanisms for the prevention of damage to DNA in spores ofBacillus species. Annu. Rev. Microbiol. 49, 29e54.

Setlow, P., 2003. Spore germination. Curr. Opin. Microbiol. 6, 550e556.Song, F., Zhang, J., Gu, A., Wu, Y., Han, L., He, K., Chen, Z., Yao, J., Hu, Y., Li, G.,

Huang, D., 2003. Identification of cry1I-type genes from Bacillus thuringiensisstrains and characterization of a novel cry1I-type gene. Appl. Environ. Micro-biol. 69, 5207e5211.

Soukas, A., Cohen, P., Socci, N.D., Friedman, J.M., 2000. Leptin-specific patterns ofgene expression in white adipose tissue. Genes Dev. 14, 963e980.

Swiecicka, I., Mahillon, J., 2006. Diversity of commensal Bacillus cereus sensu latoisolated from the common sow bug (Porcellio scaber, Isopoda). FEMS Microbiol.Ecol. 56, 132e140.

Takami, H., Han, C.G., Takaki, Y., Ohtsubo, E., 2001. Identification and distribution ofnew insertion sequences in the genome of alkaliphilic Bacillus halodurans C-125. J. Bacteriol. 183, 4345e4356.

Tusher, V.G., Tibshirani, R., Chu, G., 2001. Significance analysis of microarraysapplied to the ionizing radiation response. Proc. Nat. Acad. Sci. 98, 5116e5121.

Van Gelder, R.N., Von Zastrow, M.E., Yool, A., Dement, W.C., Barchas, J.D.,Eberwine, J.H., 1990. Amplified RNA synthesized from limited quantities ofheterogeneous cDNA. Proc. Nat. Acad. Sci. 87, 1663e1667.

Van Melis, C.C.J., Groot, N., Tempelaars, M.H., Moezelaar, R., Abee, T., 2011. Char-acterization of germination and outgrowth of sorbic acid-stressed Bacillus

D. Bassi et al. / Food Microbiology 55 (2016) 73e85 85

cereus ATCC 14579 spores: phenotype and transcriptome analysis. FoodMicrobiol. 28, 275e283.

Vepachedu, V.R., Setlow, P., 2007. Role of SpoVA proteins in release of dipicolinicacid during germination of Bacillus subtilis spores triggered by dodecylamine orlysozyme. J. Bacteriol. 189, 1565e1572.

Wang, S.T., Setlow, B., Conlon, E.M., Lyon, J.L., Imamura, D., Sato, T., Setlow, P.,Losick, R., Eichenberger, P., 2006. The forespore line of gene expression in Ba-cillus subtilis. J. Mol. Biol. 358, 16e37.

Wang, G., Zhang, P., Setlow, P., Li, Y.Q., 2011. Kinetics of germination of wet-heat-treated individual spores of Bacillus species, monitored by Raman

spectroscopy and differential interference contrast microscopy. Appl. Environ.Microbiol. 77, 3368e3379.

Wecke, T., Veith, B., Ehrenreich, A., Mascher, T., 2006. Cell envelope stress responsein Bacillus licheniformis: integrating comparative genomics, transcriptionalprofiling, and regulon mining to decipher a complex regulatory network. J.Bacteriol 188, 7500e7511.

Yakhnin, A.V., Yakhnin, H., Babitzke, P., 2008. Function of the Bacillus subtilis tran-scription elongation factor NusG in hairpin-dependent RNA polymerasepausing in the trp leader. Proc. Nat. Acad. Sci. 105, 16131e16136.