Phylogenetic and toxinogenic characteristics of Bacillus cereus group members isolated from spices and herbs

Food Control xxx (2016) 1e9

Contents lists available at ScienceDirect

Food Control journal homepage: www.elsevier.com/locate/foodcont

Phylogenetic and toxinogenic characteristics of Bacillus cereus group members isolated from spices and herbs Hendrik Frentzel*, Britta Kraushaar, Gladys Krause, Dorina Bodi, Heidi Wichmann-Schauer, Bernd Appel, Anneluise Mader German Federal Institute for Risk Assessment, Department Biological Safety, Max-Dohrn-Str. 8-10, 10589, Berlin, Germany

a r t i c l e i n f o

a b s t r a c t

Article history: Received 27 September 2016 Received in revised form 8 December 2016 Accepted 14 December 2016 Available online xxx

The foodborne pathogen Bacillus (B.) cereus is a common contaminant in spices and herbs. To further characterise B. cereus and its closely related group members present in spices and herbs, we analysed presumptive B. cereus strains isolated from six different condiments with view to B. cereus group species, phylogenetic affiliation and toxinogenic potential. Of a total of 59 isolates 44 were identified as B. cereus sensu stricto (s.s.), four as B. toyonensis-like, five as B. thuringiensis, one as B. weihenstephanensis, two as B. pseudomycoides/B. mycoides and three as undefined B. cereus group species. A maximum of three different species occurred simultaneously in the same spice sample. The isolates comprised 33 multilocus (ML) sequence types (STs), which can be assigned to three different phylogenetic groups. Except two B. pseudomycoides/B. mycoides strains, all isolates were able to produce enterotoxins and one strain the emetic toxin cereulide as detected by an immunoassay and LC-MS, respectively. The prevalence of toxin genes was 96.6% for nheA, 94.9% for hblD, 50.8% for cytK-2 and 1.7% for ces. The emetic strain was characterised by ST 869, which for the first time was assigned to an emetic B. cereus (s.s.) strain and is not part of the previously known two emetic MLST clusters. Our results demonstrate that not only B. cereus (s.s.) but also toxin producing B. thuringiensis, B. weihenstephanensis and B. toyonensis-like strains could be detected in condiments. For some isolates MLST revealed disagreements between phylogenetic relationship and the classification as B. weihenstephanensis and B. mycoides based on previously described species markers. © 2016 Elsevier Ltd. All rights reserved.

Keywords: Spore B. thuringiensis Toxin Cereulide MLST Phylogenetic tree

1. Introduction The Bacillus (B.) cereus group, also referred to as B. cereus sensu lato (s.l.) or presumptive B. cereus, comprises Gram-positive spore forming bacteria of the species B. cereus sensu stricto (s.s.), B. weihenstephanensis, B. thuringiensis, B. mycoides, B. pseudomycoides, B. anthracis, B. cytotoxicus and B. toyonensis. Due to their close genetic relationship the individual group species are difficult to distinguish. Standard methods for the detection and enumeration of presumptive B. cereus are not discriminative. Thus, routine diagnostics as well as previous studies on the occurrence of B. cereus (s.s.) in food do not consistently differentiate individual B. cereus group species. Consequently, also the contribution of the different species to foodborne disease outbreaks is uncertain.

* Corresponding author. E-mail address: [email protected] (H. Frentzel).

Moreover, based on increasing whole genome sequence data the B. cereus group taxonomy is currently under revision and might turn out much more complex: 30 rather than eight species were recently suggested (Liu et al., 2015). In our study we focus on those B. cereus group species that are validly published at the time of writing this report (see above). However, in consideration of putative taxonomic revisions and methodical difficulties of species determination, the assignment to a B. cereus group species has currently only a preliminary character. Thus, species names in our report address the most likely species based on colony morphology and the results of real-time polymerase chain reaction (PCR), microscopy and multilocus sequence typing (MLST). The uncertainty regarding the species applies to our own data as well as cited data of previous reports. Due to the ubiquitous nature and the resistance of its endospores members of the B. cereus group are detectable in many kinds of food, among them frequently in spices and dried herbs (EFSA, 2013). Not uncommonly, contamination levels of 103 to 105cfu/g

http://dx.doi.org/10.1016/j.foodcont.2016.12.022 0956-7135/© 2016 Elsevier Ltd. All rights reserved.

Please cite this article in press as: Frentzel, H., et al., Phylogenetic and toxinogenic characteristics of Bacillus cereus group members isolated from spices and herbs, Food Control (2016), http://dx.doi.org/10.1016/j.foodcont.2016.12.022

2

H. Frentzel et al. / Food Control xxx (2016) 1e9

are reported for B. cereus (s.s.) in condiments (Banerjee & Sarkar, 2003; EFSA, 2005; Hariram & Labbe, 2015; Kneifel & Berger, 1994; Sagoo et al., 2009). Also a number of foodborne disease outbreaks caused by B. cereus (s.s.) could be associated with spices (EFSA, 2016; WHO, 2014). B. cereus (s.l.) related foodborne disease is caused by enterotoxins or an emetic toxin resulting in a diarrheal or an emetic type of illness. In general, the enterotoxins are formed in the human gut after consumption of cells or spores in quantities of usually more than 105 cfu/g of food (EFSA, 2016). In contrast, the emetic toxin is already produced in the food, while the onset of toxin production is also associated with high cell densities of 105 to 106 cfu/g (Ceuppens et al., 2011). Such cell concentrations may occur in food with preferable growth conditions after addition of contaminated seasonings, especially if these foods are cooled improperly after heat treatment. Three different enterotoxins are recognised as the main cause for B. cereus (s.l.) related diarrhoea: the protein-complexes haemolysin BL (Hbl) and non-haemolytic enterotoxin (Nhe) and the single protein cytotoxin K (CytK) (Ceuppens et al., 2011). For CytK the two variants CytK-1 and CytK-2 are described with the first one showing higher cytotoxicity and being specific for B. cytotoxicus (Fagerlund, Ween, Lund, Hardy, & Granum, 2004; Guinebretiere et al., 2013). The causative agent for the emetic symptom, cereulide, is a heat stable cyclic peptide generated by the nonribosomal cereulide synthetase (Ehling-Schulz & Vukov 2005). The enterotoxin components are encoded in chromosomally located operons comprising hblC, hblD, hblA for Hbl (L2, L1 and B component) or nheA, nheB, nheC for Nhe (A, B and C component), and the single gene cytK-1 or cytK-2 for CytK1 or CytK2. The nonribosomal cereulide synthetase is encoded in the megaplasmid located cesHPTABCD genes cluster (Ceuppens et al., 2011). The ability to produce cereulide is so far reported for B. cereus (s.s.) and rarely for B. weihenstephanensis strains (Thorsen et al., 2006). In contrast, one or more enterotoxin genes (nhe, hbl and/ or cytK genes) can be found in strains of all B. cereus group species (EFSA, 2014; Guinebretiere et al., 2010). Nevertheless, cytotoxicity is not evenly distributed among the different species and different phylogenetic groups (Guinebretiere et al., 2010). Regarding B. cereus group isolates from spices and herbs data on the toxinogenic potential are limited and even more concerning phylogenetic information. However, these are important characteristics for assessing the risk associated with the occurrence of B. cereus group species in condiments or other food. Moreover, phylogenetic information could reveal links between phenotype and genotype and may further assist in epidemiological investigations. The aim of our study was to enumerate and to characterise the B. cereus group population in spices and herbs in terms of species identification, phylogenetic affiliation and toxinogenic potential. Therefore, we analysed eight condiments on the presence of B. cereus group species. In total, 59 isolates were characterised through cultural, PCR based and microscopic determination of the B. cereus group species. In addition, MLST was applied to phylogenetically classify our strains and to assist in species identification. Further, we investigated the toxin gene profiles and the actual capability to produce toxins. 2. Material and methods 2.1. Control strains As controls for the species determination and the toxinogenic analysis the following reference strains were used: B. cereus (s.s.) strain DSM 31, emetic B. cereus (s.s.) strain DSM 4312,

B. thuringiensis strain DSM 2046, B. pseudomycoides strain DSM 12442, B. weihenstephanensis strain DSM 11821, B. cytotoxicus strain DSM 22905 and B. anthracis strain 081101RA0367-13/01/2014 provided by the German Federal Research Institute for Animal Health (FLI). 2.2. Enumeration and isolation of presumptive B. cereus in spices and herbs In order to quantify and isolate members of the B. cereus group in condiments the following representatives were analysed: allspice, cinnamon, nutmeg, paprika and pepper for spices and basil, oregano and parsley for herbs. All spices and herbs were provided for research purposes by FUCHS Gewürze GmbH, Dissen, Germany as dried and ground powders. Except oregano and parsley all matrices were processed by heat steam treatment by the manufacturer for microbiological decontamination. For presumptive B. cereus enumeration 10 g triplicates of each matrix were dissolved in 190 ml peptone water (0.1%, w/v; Merck, Darmstadt, Germany) and shaken by hand for 1 min. Using peptone water tenfold dilutions were prepared and 100 ml of the dilutions were plated on polymyxin pyruvate egg yolk mannitol bromothymol blue agar (PEMBA; Oxoid, Wesel, Germany) in duplicate. Typical colonies were counted after incubation at 37
C for 24 h to determine the cfu/g. If no typical colonies were countable, samples were again diluted as described above and 1 ml of the first dilution was split plated on three PEMBA plates in duplicate. The theoretical limit of detection (LOD) was therefore 20 cfu/g. Randomly picked typical colonies were confirmed on mannitol egg yolk polymyxin agar (MYP; Merck) and sheep blood agar (SBA; Mast Diagnostica, Reinfeld, Germany) plates. Confirmed strains were kept at 80
C in glycerine culture. Calculation and declaration of cfu/g was carried out in accordance with ISO 7218:2014-09 (Anonymous, 2014a). 2.3. Species identification within the B. cereus group Species determination was based on colony morphology, PCR, microscopy and MLST. For the attribution of isolates to individual species we made use of the following parameters: B. mycoides ¼ rhizoid growth (B. mycoides may in addition also be PCR positive for motB using MotB_1 or MotB_2 probe) (OliwaStasiak, Kolaj-Robin, & Adley, 2011); B. pseudomycoides ¼ rhizoid growth and PCR positive only for bpm; B. weihenstephanensis ¼ nonrhizoid growth and PCR positive only for motB using the MotB_2 probe (Oliwa-Stasiak et al., 2011); B. anthracis ¼ PCR positive only for motB using the MotB_1 probe and for PL3 (Oliwa-Stasiak et al., 2011; Wielinga et al., 2011); B. thuringiensis ¼ PCR positive only for motB using the MotB_1 probe and for cry1 or microscopic positive for parasporal crystals (Oliwa-Stasiak et al., 2011; Wielinga et al., 2011); B. cytotoxicus ¼ PCR positive only for cytK1 (Guinebretiere, Fagerlund, Granum, & Nguyen-The, 2006, 2013); B. cereus (s.s.) PCR positive only for motB using the MotB_1 probe and exclusion of alternative B. cereus group species based on the parameters mentioned above and the results of MLST (below). The species B. toyonensis is formally only represented by the type strain NCIMB 14858T (previously BCT-7112T). This strain was so far only differentiated from B. cereus (s.s.) by digital DNA:DNA hybridization (dDDH) based on whole genome sequences and supported by MLST using the housekeeping genes adk, ccpA, glpT, pyrE, recF and sucC (Jimenez et al., 2013). Both of these datatypes whole genome sequences and MLST using the outlined genes were not generated for our isolates. Instead, we made use of the publically available sequence types (STs) (http://pubMLST.org/ bcereus/(Jolley & Maiden, 2010)) of NCIMB 14858T as well as three other strains belonging to the same genomospecies (VD148,

Please cite this article in press as: Frentzel, H., et al., Phylogenetic and toxinogenic characteristics of Bacillus cereus group members isolated from spices and herbs, Food Control (2016), http://dx.doi.org/10.1016/j.foodcont.2016.12.022

H. Frentzel et al. / Food Control xxx (2016) 1e9

VD115 and MC28) (Jimenez et al., 2013). By incorporation of these STs in a phylogenetic tree the relationship to our isolates was shown based on MLST using seven different housekeeping genes (see subsection 2.4). For the application of PCR as well as MLST analysis, DNA was extracted from 500 ml of an overnight culture (37
C) of an isolate in brain heart infusion broth (BHI; Oxoid) using the DNeasy Blood & Tissue Kit according to the manual (Qiagen, Hilden, Germany). Primers and probes were used as previously described by Guinebretiere et al. (2006), Oliwa-Stasiak et al. (2011), Wielinga et al. (2011) and modifications according to Frentzel et al. (this issue). Primers and probes were synthesized by Eurofins Genomics, Ebersberg, Germany and their sequences are given in Table 1. In order to prepare the microscopic examination for parasporal crystals, isolates were streaked on nutrient agar (Sifin, Berlin, Germany) supplemented with 40 mg/l MnSO4 and 100 mg/l CaCl2 and incubated for 48 h at 30
C. Afterwards, a colony fragment was mixed with 5 ml of PBS on a microscope slide and analysed using a 100 phase contrast objective.

3

HS Mix (Bioline, Luckenwalde, Germany), 1.0 mM of each primer and 5 ml of the 1:10 diluted DNA extract. The amplification program was composed of an initial denaturation of 5 min at 95
C and 30 cycles of 30 s at 94
C, 30 s at the specific primer annealing temperature and 30 s at 72
C with a final extension of 5 min at 72
C. Success of the PCR was verified by agarose gel electrophoresis. MLST gene PCR products were purified using the illustra GFX PCR DNA and Gel Band Purification Kit according to the manual (GE HealthcareEurope, Freiburg, Germany). Sequencing of PCR products was accomplished by Eurofins Genomics. Obtained DNA sequence data was processed and analysed using Accelrys Gene 2.5 software. For the attribution of sequence types (STs), sequence data was submitted to the MLST database and new alleles and STs were nominated by MLST database curators. Using the MEGA7 software (http://www.megasoftware.net/) a Neighbour Joining Tree (Maximum likelihood; 1000 x bootstrap) was constructed from concatenated sequences of the seven housekeeping gene fragments of our isolates and other representative sequences (Jimenez et al., 2013; Priest, Barker, Baillie, Holmes, & Maiden, 2004).

2.4. Multilocus sequence typing 2.5. Detection of toxin genes and toxin production This publication made use of the Bacillus cereus Multilocus Sequence Typing website (http://pubMLST.org/bcereus/) developed by Keith Jolley and sited at the University of Oxford (Jolley & Maiden, 2010). Fragments of seven housekeeping genes (glpF, gmK, ilvD, pta, pur, pycA and tpi) were amplified by PCR (Veriti thermal cycler, Thermo Fisher Scientific, Waltham, USA) using primers as specified in the MLST database. Suggested alternative primers outlined in the MLST database were used, if original primers yielded no product. Each 50 ml PCR reaction contained 20 ml MyTaq™

The detection of the representative toxin genes nheA, hblD and a specific part of the ces genes was based on real-time PCR as previously described by Wehrle, Didier, Moravek, Dietrich, and Martlbauer (2010) with slight modifications: Each 25 ml reaction contained 12.5 ml 2x SensiFAST™ HRM Kit (Bioline), 0.2 mM of primers mp3L1R1for and mp3L1R1rev, 0.2 mM of primers mp3AR2for and mp3AR2rev and 0.25 mM of primers ces_SYBR_F and ces_SYBR_R, and 5 ml of the target DNA. The amplification

Table 1 Sequences of primers and probes used in this study. Name

Sequence (50 e30 )a

Target

Reference

BCFomp2 BCRomp2 MotB_1 MotB_2

CGCCTCGTTGGATGACG GATATACATTCACTTGACTAATACCG FAM-TTCAAGCATCTTTGACAATTTTACTGCAT-BBQ HEX-TTCAAGCATCTTYGATAATTTTACTGTAT-BBQ (Y ¼ T/C)

motB

Oliwa-Stasiak et al. (2011)

BpmF BpmR2 Bpm_1

TAATTTAGGGGGGCATCTTTACTTTTC CTATACCCAAAACTTAGATATGCTC TxRed-CTGAGAAGGTAGTCATACGCTATACATG-BBQ

bpm

Oliwa-Stasiak et al. (2011)

Btpri_f Btpri_r Tqpro_Bt

GCAACTATGAGTAGTGGGAGTAATTTAC TTCATTGCCTGAATTGAAGACATGAG Cy5-ACGTAAATACACTTGATCCATTTGAAAAG-BHQ2

cry1

Wielinga et al. (2011)

PL3_f PL3_r Tqpro_PL3

AAAGCTACAAACTCTGAAATTTGTAAATTG CAACGATGATTGGAGATAGAGTATTCTTT ROX-AACAGTACGTTTCACTGGAGCAAAATCAA-BHQ1

Pl3

Wielinga et al. (2011)

mp3L1R1for mp3L1R1rev

AGTTATTGCAGCTATTGGAGG GTCCATATGCTTAGATGCTGTGA

hblD

Wehrle et al. (2010)

mp3AR2for mp3AR2rev

TTCAAATTCAAAAGAATGTTGAAGAAGG GATTTGTTTGCTTATTCATTTCATCAC

nheA

Wehrle et al. (2010)

ces_SYBR_F ces_SYBR_R

CACGCCGAAAGTGATTATACCAA CACGATAAAACCACTGAGATAGTG

ces

Fricker et al. (2007)

CK1F CK1R

ACAGATATCGG(G,T)CAAAATGC TCCAACCCAGTT(A,T) (GC)CAGTTC

cytK-1

Guinebretiere et al. (2006)

CK2F CK2R

CAATCCCTGGCGCTAGTGCA GTGIAGCCTGGACGA AGTTGG

cytK-2

Guinebretiere et al. (2006)

a Probes are labelled with reporter dyes (50 -end) and quenchers (30 -end): FAM, 6-carboxyfluorescein; HEX, hexachlorofluorescein; TxRed, Texas Red; Cy5, cyanine5; BBQ, BlackBerry Quencher, BHQ, Black Hole Quencher; ROX 6-carboxyl-x-rhodamine.

Please cite this article in press as: Frentzel, H., et al., Phylogenetic and toxinogenic characteristics of Bacillus cereus group members isolated from spices and herbs, Food Control (2016), http://dx.doi.org/10.1016/j.foodcont.2016.12.022

4

H. Frentzel et al. / Food Control xxx (2016) 1e9

program carried out on a Veriti thermal cycler (Thermo Fisher Scientific) consisted of an initial denaturation at 95
C for 10 min, followed by 35 cycles at 95
C for 15 s, 59
C for 10 s and 72
C for 20 s, subsequent heating to 95
C for 20 s, cooling to 60
C for 20 s and reheating to 95
C at a rate of 0.1
C/s. The detection of the cytK-1 and cytK-2 genes followed the provisions of Guinebretiere et al. (2006). Primers were synthesized by Eurofins Genomics and their sequences are given in Table 1. Besides for the possession of toxin genes, isolates were examined for actual toxin production of the enterotoxins Hbl and Nhe, and the emetic toxin cereulide (only ces positive strains). For enterotoxin testing the GLISA Duopath® Cereus Enterotoxins (Gold Labelled Immuno Sorbent Assay, Merck) was used according to the manual. If no toxin production was visible after 4 h of incubation as suggested in the manual, isolates were tested again after overnight incubation. In order to extract cereulide from cells, previously published €ggblom et al. (2002) were protocols of Marxen et al. (2015) and Ha considered and modified as follows: For cereulide production 100 ml of an overnight culture grown in tryptic soy broth (TSB; Oxoid) was inoculated to 50 ml TSB in a 300 ml flask followed by incubation at room temperature, shaking at 150 rpm for 48 h. After incubation, the liquid cultures were transferred into 50 ml Falcon tubes and centrifuged at 3800 rpm for 10 min. The supernatant was decanted and the resulting pellet was frozen and thawed three times to destabilize the cell wall. For cereulide extraction from the cells the pellet was dissolved in 10 ml methanol and shaken at 250 rpm overnight at room temperature. The resulting extract was centrifuged at 3800 rpm for 10 min and the supernatant was membrane-filtered (pore size 0.2 mm, polyethersulfone membrane; VWR, Radnor, USA). Until further analyses the extracts were stored at 4
C. The cereulide production was analysed in accordance with ISO 18465:2014e11 e Draft (Anonymous, 2014b). Cereulide analysis was performed using a TSQ Quantiva triple quadrupole mass spectrometer coupled to a UHPLC (Thermo Fisher Scientific). Chromatographic separation was achieved on a Thermo Hypersil Gold C18 1.9 mm, 150  2.1 mm, column (Thermo Fisher Scientific) at 40
C column temperature using a gradient program. Solvent A of the gradient contained 0.1% formic acid and 5 mM ammonium formate in water and solvent B consisted of 0.1% formic acid and 5 mM ammonium formate in methanol. The flow rate was maintained at 400 ml/min. Starting conditions of the gradient were 60% B for 30 s, linearly increased to 98% B for 18 s and kept at 98% for 3.2 min, then decreased to 60% within 6 s and finally kept at 98% B for 1.9 min. The injection volume was 1 ml. The conditions of electrospray positive ionisation (ESIþ) were: 3.5 kV ion spray voltage, 333
C ion transfer tube temperature, 400
C vaporizer temperature and gas flow rates of 40 psi sheath gas, 1 psi sweep gas and 10 psi auxiliary gas. For cereulide identification the product ions m/z 1170.83 > 172.33 (collision energy (CE) 58.6 V), 1170.83 > 314.28 (CE 58.6 V), 1170.83 > 357.16 (CE 58.1 V) and 1170.83 > 499.32 (CE 58.6 V) were used. The latter was used for quantification. Signal ratios of these product ions were also checked to be in the same range in the samples and the standard solutions as an identification criterion. For calibration a standard curve was set up by diluting cereulide stock solution (Chiralix B. V., Nijmegen, The Netherlands) in methanol resulting in final concentrations of 0.4, 1.6 and 8 ng/ml. 3. Results and discussion 3.1. Presumptive B. cereus group species in spices and herbs Presumptive B. cereus counts in the eight analysed spices and

herbs ranged from <80 cfu/g in basil, paprika and pepper to 1.6  103 cfu/g in allspice (Table 2). In samples of cinnamon and nutmeg no presumptive B. cereus could be detected (LOD ¼ 20 cfu/ g). These findings are in concordance with previous studies by Banerjee and Sarkar (2003), Hariram and Labbe (2015), Kneifel and Berger (1994) and Sagoo et al. (2009) reporting low quantities (<104 cfu/g) of B. cereus (s.s.) in the majority of analysed condiment samples. However, in all of these studies, except Hariram and Labbe (2015), also maximum concentrations of more than 105cfu/g are described. It should be noted that, except one Process Hygienic Criterion for dried infant formula and dried dietary foods for special medical purposes, no microbiological criteria for presumptive B. cereus in food are applied in terms of Commission Regulation (EC) No 1441/ 2007 of 5 December 2007 on microbiological criteria for foodstuffs (Anonymous, 2007). However, EFSA considers B. cereus (s.s.) concentrations of 105 cfu/g of food in general (rarely also lower concentrations of 102 - 103 cfu/g) as potential cause of foodborne disease (EFSA, 2016). Interestingly, also for B. thuringiensis this contamination level is considered as a risk for consumers by EFSA (EFSA, 2016). Specific for dried spices and herbs the German Society for Hygiene and Microbiology (DGHM) advises a warning value of 104 cfu/ g for presumptive B. cereus (DGHM, 2011). As the highest presumptive B. cereus count in our samples was 103 cfu/g, all analysed condiments meet these food safety recommendations. 3.2. Species identification Of a total of 59 isolates randomly picked from six matrices we identified 44 as B. cereus (s.s.), four as B. toyonensis-like, five as B. thuringiensis, one as B. weihenstephanensis, two as B. pseudomycoides/B. mycoides and three as undefined presumptive B. cereus (for details see below, Fig. 1 and Table 3). At the most, three different species were present simultaneously in the same matrix: B. cereus (s.s.), B. weihenstephanensis and B. thuringiensis in pepper. In contrast, only one species (B. thuringiensis) could be detected in paprika. Notably, one isolate (BfR-BA-475) showed the same ST as the B. thuringiensis strain ABTS-1857, which is applied in the commercial insecticidal formulation Xentari®. Initially 48 of 59 isolates were identified as B. cereus (s.s.) based on the exclusion of B. pseudomycoides, B. mycoides, B. weihenstephanensis, B. anthracis, B. thuringiensis and B. cytotoxicus by means of cultural, PCR and microscopic approaches (see subsection 2.3). However, four of the 48 putative B. cereus (s.s.) strains (BfR-BA336, 338, 364 and 365) branched within a clearly defined B. toyonensis cluster in the phylogenetic tree based on MLST (Fig. 1). Besides our isolates, this cluster comprised the B. toyonensis type strain (NCIMB 14858T) and three more strains of the B. toyonensis genomospecies (MC28, VD 115 and VD 148; Jimenez et al., 2013), for which STs were available in the MLST database. For this reason we assume that these four isolates could as well belong to the B. toyonensis species and refer to them as B. toyonensis-like-strains. Five of the 59 isolates were identified as B. thuringiensis strains mainly on the basis of microscopy as the cry1 specific primers only gave a positive PCR signal for one of the isolates. Also Hariram and Labbe (2015) differentiated B. cereus (s.s.) from B. thuringiensis in 88 isolates from various spices and herbs. In their study a ratio of 12.5% B. thuringiensis to B. cereus (s.s.) was found, which is similar to our results (10.2%) as well as previous findings for rice (11.7%) (Ankolekar, Rahmati, & Labbe, 2009), milk (8.9%) (Zhou, Liu, He, Yuan, & Yuan, 2008) or dairy associated samples (4.3%) (Cui et al., 2016). In contrast, Frederiksen, Rosenquist, Jorgensen, and Wilcks

Please cite this article in press as: Frentzel, H., et al., Phylogenetic and toxinogenic characteristics of Bacillus cereus group members isolated from spices and herbs, Food Control (2016), http://dx.doi.org/10.1016/j.foodcont.2016.12.022

H. Frentzel et al. / Food Control xxx (2016) 1e9

5

Table 2 Presumptive B. cereus detected in spices and herbs.

Allspice Basil Cinnamon Nutmeg Oregano Paprika Parsley Pepper a b c d

cfu/ga ± standard deviation

Number of isolatesb

Number of STsc

1.6  103 ± 2.0  102 Present, <8  101
12 14 0 0 14 3 13 3

10 7 0 0 8 2 4 3

Colony forming units per gram of matrix of three parallel samples, based on counts of typical colonies for presumptive B. cereus on PEMBA. Isolates were obtained from presumptive B. cereus colonies randomly picked from PEMBA plates. Number of different sequence types of all isolates within one matrix. DL ¼ detection limit (20 cfu/g).

(2006) found a much higher ratio of B. thuringiensis to B. cereus (s.s.) of 39.1% in 128 isolates from fresh vegetables and fruits. Likewise, Rosenquist, Smidt, Andersen, Jensen, and Wilcks (2005) reported a proportion of 77.5% in 40 isolates randomly chosen from numerous ready-to-eat foods. Based on non-rhizoid growth and an exclusive positive PCR signal originating from the MotB_2 probe four isolates were initially classified as B. weihenstephanensis. However, three of them (BfR-BA343, 363 and 460) branched close to B. cereus (s.s.) in clade 1 distant to other B. weihenstephanensis, which are usually found close to B. mycoides in clade 3 according to the classification of Priest et al. (2004). Yet, the close relationship between B. weihenstephanensis and B. mycoides was recently confirmed by calculating intergenomic distances based on 224 whole genome sequences (Liu et al., 2015). Therefore, these three putative B. weihenstephanensis strains may rather be treated as B. cereus (s.s.) strains. Due to the contradiction between PCR result and MLST we refer to these isolates as undefined presumptive B. cereus. Consequently, the MotB_2 probe used to distinguish B. weihenstephanensis from B. cereus (s.s.) might sometimes also target PCR amplificates deriving from B. cereus (s.s.) strains although it was found to be discriminative by Oliwa-Stasiak et al. (2011). In contrast, the fourth isolate initially classified as B. weihenstephanensis was found in the expected clade 3. Therefore, we kept the classification as B. weihenstephanensis for this strain. Based on rhizoid growth and no PCR signal for any of the targets two isolates (BfR-BA-371 and 472) were classified initially as B. mycoides strains. In conflict to this, these isolates were found in close proximity to B. pseudomycoides, which branched in distance to B. mycoides and B. weihenstephanensis in the MLST based phylogenetic tree. This phylogenetic distance is in concordance with the findings of Liu et al. (2015). Hence, we assume that these two isolates represent more likely B. pseudomycoides strains and refer to them as B. pseudomycoides/B. mycoides. As a consequence it can be stated that the applied primers targeting the bpm sequence might miss some B. pseudomycoides strains. Indeed, even in the original study of Oliwa-Stasiak et al. (2011) one of eleven B. pseudomycoides strains (WS 3120) was not detected using bpm specific primers. In summary, four different species (B. cereus (s.s.), B. weihenstephanensis, B. thuringiensis and B. pseudomycoides/B. mycoides) as well as B. toyonensis-like strains were identified in 59 isolates derived from six spice and herb matrices. 3.3. Phylogenetic characteristics based on MLST The STs obtained by MLST analysis for all isolates are given in Table 3 and a phylogenetic tree, generated from concatenated sequences of the seven MLST loci, is presented in Fig. 1.

All together 33 different STs were assigned to 59 isolates. Of these 33 STs, ten were already existing in the MLST database (STs 12, 15, 23, 212, 223, 614, 760, 869, 1133, 1265). Strains of these known STs originate from diverse isolation sources such as food, soil, leaves, environmental samples or sediments. In our study different spice and herb isolation sources mostly went along with diverging STs. However, the same ST (ST 23) was present in samples of oregano and basil. Hence, STs need not be linked to isolation sources. In our study different species were always represented by different STs. Though, ST 23, assigned to B. cereus (s.s.) strains in our study, can also represent B. thuringiensis strains (Priest et al., 2004). The same applies for ST 12, which we assigned to a B. cereus (s.s.) strain, but was designated to B. thuringiensis strains by Priest et al. (2004). Likewise, Cardazzo et al. (2008) reported identical STs (e.g. ST 12) for different subspecies of B. thuringiensis and B. cereus (s.s.) isolates. In agreement with this, most of our B. cereus (s.s.) isolates as well as B. thuringiensis strains were distributed over all linages (Tolworthi, Kurstaki, Thuringiensis and Sotto) in clade 2 of the phylogenetic tree referring to the classification of Priest et al. (2004). Clade two was meant to be dominated by B. thuringiensis in the original study (Priest et al., 2004), but was later described to frequently include B. cereus (s.s.) as well (Hoffmaster et al., 2008; Zahner, Silva, de Moraes, McIntosh, & de Filippis, 2013). In summary, our results underline previous findings that, based on MLST, B. cereus (s.s.) and B. thuringiensis cannot be differentiated. Indeed, Liu et al. (2015) reported recently that even by comparing intergenomic distances based on whole genome sequences B. cereus (s.s.) and B. thuringiensis could not be resolved as two distinct species. The same applied to B. weihenstephanensis and B. mycoides, which, according to the authors, should be treated as one species. At the same time 22 to 23 putative novel species were predicted including the recently suggested (not yet validly published) novel B. cereus group species B. gaemokensis (Jung et al., 2010), B. manliponensis (Jung et al., 2011) and B. bingmayongensis (Liu et al., 2014). Seven of our isolates were found in clade 1 of the Priest scheme. One of these isolates is the emetic strain BfR-BA-331. All emetic strains previously characterised by MLST can be found in clade 1. Emetic B. cereus (s.s.) are described as a closely related group, hitherto only represented by four different STs, which are ST 26, 144, 164 and 165 (Ehling-Schulz & Svensson 2005; Vassileva et al., 2007). However, within this group STs 26 and 165 as well as STs 144 and 164 form two distinct clusters (Vassileva et al., 2007). Isolate BfR-BA-331 is characterised by ST 869, which branches separate from both of these clusters. ST 869 is also designated to a Kenyan soil isolate in the MLST database but this isolate is not further described. To our knowledge, this is the first report assigning ST 869

Please cite this article in press as: Frentzel, H., et al., Phylogenetic and toxinogenic characteristics of Bacillus cereus group members isolated from spices and herbs, Food Control (2016), http://dx.doi.org/10.1016/j.foodcont.2016.12.022

6

H. Frentzel et al. / Food Control xxx (2016) 1e9

(2010). This group comprises B. cereus (s.s.) and B. thuringiensis strains considered as moderate to high cytotoxic (Ceuppens et al., 2011; Guinebretiere et al., 2010). Clade 1, including seven of our isolates, represents the phylogenetic group III of Guinebretiere et al. (2010). This group comprises B. cereus (s.s.) (including emetic strains) as well as B. thuringiensis strains of low and high cytotoxicity and B. anthracis strains of high cytotoxicity. Finally, clade 3, harbouring three of our isolates, corresponds to group VI of Guinebretiere et al. (2010), which consists of B. weihenstephanensis, B. mycoides and B. thuringiensis strains of low cytotoxicity. With regard to enterotoxin genes (see also subsection 3.4) we found that all isolates of the same ST were characterised by the same toxin gene profile. However, based on only six STs that include more than one isolate this finding cannot be generalised. 3.4. Toxin genes and toxin production

Fig. 1. Phylogenetic tree (Neighbour Joining, 1000 bootstrap) constructed from concatenated sequences of seven housekeeping gene fragments including isolates of this study and other representative sequences from the MLST database. Clades and lineages are indicated as described by Priest et al. (2004) with the addition of a Toyonensis-cluster. Bc: B. cereus, Bm: B. mycoides, Bp: B. pseudomycoides, Bt: B. thuringiensis, Bw: B. weihenstephanensis.

to an emetic B. cereus (s.s.) strain. Hence, we add a fifth distinct ST to the previously known four STs characteristic for emetic B. cereus (s.s.). Three of our isolates, one B. weihenstephanensis strain and the two B. pseudomycoides/B. mycoides strains, were found in clade 3. As mentioned above most of our strains cluster in clade 2. According to Tourasse et al. (2011) clade 2 in the Priest scheme can be translated to phylogenetic group IV in terms of Guinebretiere et al.

To characterise the toxinogenic potential of our isolates we investigated the presence of toxin genes by PCR using nheA, hblD, ces, cytK-1 and cytK-2 as representative genes. Subsequently we examined the actual toxin production capabilities with regard to Nhe, Hbl and cereulide applying the GLISA Duopath® and mass spectrometry, respectively (ISO 18465:2014e11 e Draft). Toxin gene profiles and toxin production capabilities are presented in Table 3. With the exception of the two B. pseudomycoides/B. mycoides strains all isolates carried at least two toxin genes. One of the isolates carried the ces sequence, thus, being a potentially emetic strain (BfR-BA-331). None of the isolates carried the cytK-1 gene, responsible for a highly toxic variant of CytK, which is specific for B. cytotoxicus (Fagerlund et al., 2004; Guinebretiere et al., 2013). In total, three different toxin gene combinations occurred: 27 isolates were positive for nheA and hblD, 29 isolates carried nheA, hblD and cytK-2 and one isolate harboured nheA, cytK-2 and ces. The prevalence of the individual toxin genes was 96.6% for nheA, 94.9% for hblD, 50.8% for cytK-2 and 1.7% for ces. No toxin genes could be detected in the two B. pseudomycoides/B. mycoides strains. The actual toxin production capabilities reflected the genetic configuration very well. Thus, all isolates carrying the nheA gene and/or hblD gene were also able to produce the respective toxin, except the B. weihenstephanensis strain (BfR-BA-476). This strain, carrying nheA and hblD, was only able to produce Nhe, and even Nhe could only be detected after overnight incubation (at 37
C as well as 30
C). Also two of the five B. thuringiensis isolates (BfR-BA475 and BfR-BA-478) needed overnight incubation times to produce detectable levels of Nhe as well as Hbl. Worth mentioning, also the four isolates belonging to the B. toyonensis cluster harboured nheA, hblD and one isolate additionally cytK-2 (BfR-BA-336) and they also produced Nhe and Hbl. While the B. toyonensis type strain NCIMB 14858T is known to possess the nhe and hbl operons, findings on its cytotoxicity are somewhat contradictory. Though some authors deem it noncytotoxic (Blanch, Mendez, Castel, & Reina, 2014; Trapecar et al., 2011), EFSA considers it as cytotoxic based on a study of Doll, Ehling-Schulz, and Vogelmann (2013) and a Technical dossier (EFSA, 2014). In addition, the MLST database includes two strains of ST 158. One of these strains is MC28 belonging to the genomospecies B. toyonensis (Jimenez et al., 2013) while the second strain, NC 543, was mentioned as an enterotoxin producer in a study of Vassileva et al. (2007). Our ces positive isolate was also able to form cereulide. Under the conditions outlined in subsection 2.5 isolate BfR-BA-331 achieved a concentration of 802.76 ng/ml cereulide. In comparison, the emetic control strain DSM 4312, originating from a food poisoning incident, yielded a cereulide concentration of 1235.71 ng/ml in our study.

Please cite this article in press as: Frentzel, H., et al., Phylogenetic and toxinogenic characteristics of Bacillus cereus group members isolated from spices and herbs, Food Control (2016), http://dx.doi.org/10.1016/j.foodcont.2016.12.022

H. Frentzel et al. / Food Control xxx (2016) 1e9

7

Table 3 Characteristics of B. cereus group strains isolated from spices and herbs. Isolate

Matrix

Species

BfR-BA331 332 336 338 339 364 365 366 367 368 369 370 347 363 460 461 462 463 464 465 466 467 468 469 470 471 333 334 335 337 343 345 356 357 358 359 360 361 362 472 473 474 475 340 341 342 344 346 348 350 351 352 353 354 355 371 476 477 478

Allspice Allspice Allspice Allspice Allspice Allspice Allspice Allspice Allspice Allspice Allspice Allspice Basil Basil Basil Basil Basil Basil Basil Basil Basil Basil Basil Basil Basil Basil Oregano Oregano Oregano Oregano Oregano Oregano Oregano Oregano Oregano Oregano Oregano Oregano Oregano Oregano Paprika Paprika Paprika Parsley Parsley Parsley Parsley Parsley Parsley Parsley Parsley Parsley Parsley Parsley Parsley Parsley Pepper Pepper Pepper

B. cereus B. cereus B. toyo.a B. toyo.a B. cereus B. toyo.a B. toyo.a B. cereus B. cereus B. cereus B. cereus B. cereus B. cereus undefinedb undefinedb B. cereus B. cereus B. cereus B. cereus B. cereus B. cereus B. cereus B. cereus B. cereus B. cereus B. cereus B. cereus B. cereus B. cereus B. cereus undefinedb B. cereus B. cereus B. cereus B. cereus B. cereus B. cereus B. cereus B. cereus B. ps/myc B. thur. d B. thur. d B. thur.d B. cereus B. cereus B. cereus B. cereus B. cereus B. cereus B. cereus B. cereus B. cereus B. thur.d B. cereus B. cereus B. ps/myc B.weihen.e B. cereus B. thur.d

Species marker

Toxin genes

Toxin production

MotB_1

MotB_2

bpm

PL3

cry1

crystals

nheA

hblD

ces

cytK-1

cytK-2

Nhe

Hbl

Cereulide

þ þ þ þ þ þ þ þ þ þ þ þ þ e e þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ e þ þ þ þ þ þ þ þ e þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ e e þ þ

e e e e e e e e e e e e e þ þ e e e e e e e e e e e e e e e þ e e e e e e e e e e e e e e e e e e e e e e e e e þ e e

e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e

e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e

e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e þ e e e e e e e e e e e e e e e e

e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e þ þ þ e e e e e e e e e þ e e e e e þ

þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ e þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ e þ þ þ

e þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ e þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ e þ þ þ

þ e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e

e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e e

þ þ þ e þ e e þ þ þ þ þ e e e þ þ e e e e e e e e þ e þ e e e e e þ e e e e e e þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ e e þ þ

þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ e þ þ (þ) þ þ þ þ þ þ þ þ þ þ þ þ e (þ) þ (þ)

e þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ þ e þ þ (þ) þ þ þ þ þ þ þ þ þ þ þ þ e e þ (þ)

þ ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND ND

ST

869 1250 1264 1257 1258 223 223 1276 1276 760 1277 1278 614 1307 1302 1279 1280 23 23 23 23 23 23 23 23 1303 12 1298 23 23 1299 1300 1306 1301 23 12 23 23 23 1309 1133 1133 15 1259 1259 1259 1259 1259 1259 1259 1265 1259 212 1259 1259 1308 1304 1281 1297

(þ) ¼ only positive after overnight incubation. ND ¼ not determined; ST ¼ sequence type. a B. toyonensis-like strains. b Undefined presumptive B. cereus. c B. pseudomycoides/B. mycoides. d B. thuringiensis. e B. weihenstephanensis.

The overall prevalence of the ces genes in B. cereus (s.s.) strains is considered to be lower than 5% (Wehrle et al., 2010) (compare Ceuppens et al., 2011, p. 1.5%), but can be much higher in food associated samples (e.g. 11% in ready-to-eat foods; Ceuppens et al.,

2011). In line with this, Kim, Baek, Lee, and Oh (2013) found 13.2% ces positive strains out of 68 isolates from red pepper. In this study ces occurred in combination with nhe genes, with nhe and hbl genes as well as nhe, hbl and cytK. In a study of Ehling-Schulz and

Please cite this article in press as: Frentzel, H., et al., Phylogenetic and toxinogenic characteristics of Bacillus cereus group members isolated from spices and herbs, Food Control (2016), http://dx.doi.org/10.1016/j.foodcont.2016.12.022

8

H. Frentzel et al. / Food Control xxx (2016) 1e9

Svensson (2005) all 24 ces positive isolates were also able to produce the toxin cereulide. None of these strains carried the hbl genes, whereas most strains (22) carried nheA and two of them additionally the cytK gene, as did our emetic strain. The overall prevalence of the enterotoxin genes for B. cereus (s.s.), and similarly for B. thuringiensis, is much higher than for the ces genes: 84e100% for nhe genes, 29e92% for hbl genes and 37e89% for cytK (Ceuppens et al., 2011). Guinebretiere, Broussolle, and Nguyen-The (2002) reports a frequency of 63% and 33% of the toxin gene combination nhe, hbl and cytK among B. cereus (s.s.) food-poisoning strains and among foodborne strains, respectively. Of the 88 isolates from U.S. retail spices in a study of Hariram and Labbe (2015), p. 82% and 72% of B. cereus (s.s.) isolates were positive for nhe and hbl genes whereas these genes could be found in 72% and 67% of B. thuringiensis isolates, respectively (no ces was detected). Most of these isolates were also able to produce the corresponding enterotoxin. Interestingly, Rivera, Granum, and Priest (2000) report an equal cytotoxicity of nheB/C and hblC/D positive B. thuringiensis strains compared to B. cereus (s.s.) strains. Our results show that different B. cereus group species -B. cereus (s.s.), B. thuringiensis, B. toyonensis-like and (less effective) B. weihenstephanensis-isolated from spices and herbs had the potential to produce enterotoxins and one isolated strain produced the emetic toxin cereulide. However, the possibility to multiply to toxin producing quantities of cells (>105 cfu/g) in foods seasoned with B. cereus (s.l.) containing spices depends on various factors. Not only the composition and treatment of the food itself may be decisive, but also the overall microbial population of the condiment, as suggested by Hariram and Labbe (2016). In their study the growth of B. cereus (s.s.) in seasoned rice was probably hampered by the competing microflora, initially present in higher concentrations in the applied pepper (aerobic mesophilic population ~ 106 cfu/g) compared to B. cereus (s.s.) (~102 cfu/g). Moreover, amounts of produced toxins are highly strain specific (Jessberger et al., 2015). In conclusion, our results show that the B. cereus group population in spices and herbs is very diverse. Within 59 isolates we found at least four different species and 33 different STs, which belong to three different phylogenetic groups (III, IV and VI) in terms of Guinebretiere et al. (2010). MLST analysis failed to link isolation source and ST and moreover to differentiate between B. cereus (s.s.) and B. thuringiensis. However, it proved useful for the classification of strains to phylogenetic groups associated with varying cytotoxicity (Ceuppens et al., 2011; Guinebretiere et al., 2010). Additionally, MLST revealed disagreements between phylogenetic relationship and species markers such as the bpm sequence or the MotB_2 target. Furthermore, an MLST-based cluster analysis may serve to distinguish B. toyonensis from B. cereus (s.s.). As stated above, the taxonomy of the B. cereus group is subject to controversial discussions as the phenotypic species identification is challenged by increasing DNA sequence data, providing a new picture of the phylogenetic relationships within the B. cereus group. Thus, the species assignment presented in our report, based on phenotypic and genotypic characteristics, might be modified in upcoming years as new B. cereus group species will likely be defined while some of the established might disappear (Liu et al., 2015). Regarding the consumer risk, the maximum contamination level of 103 cfu/g in the analysed spices and herbs should not be considered as a risk through the bare consumption of the condiments. Nevertheless, outgrowth and proliferation of B. cereus group species in seasoned food could potentially cause foodborne diseases, as (except two B. pseudomycoides/B. mycoides strains) all isolates were able to produce enterotoxins and one strain the emetic toxin cereulide.

Acknowledgement The article was prepared in the framework of the EU project SPICED (Grant Agreement: 312631; www.spiced.eu) with the financial support from the 7th Framework Programme of the EU. This publication reflects the view of the authors, and the European Commission cannot be held responsible for any use which may be made of the information contained therein. The authors would like to thank Sara Schaarschmidt and Lea Herges for their assistance in the project and Katja Drache, Ylanna Kelner-Burgos, Daniel Leeser and Wolfgang Herkt for supporting the experiments.

References Ankolekar, C., Rahmati, T., & Labbe, R. G. (2009). Detection of toxigenic Bacillus cereus and Bacillus thuringiensis spores in US rice. International Journal of Food Microbiology, 128(3), 460e466. Anonymous. (2007). COMMISSION REGULATION (EC) No 1441/2007 of 5 December 2007 amending Regulation (EC) No 2073/2005 on microbiological criteria for foodstuffs. Anonymous. (2014a). Microbiology of food and animal feeding stuffs - general requirements and guidance for microbiological examinations (ISO 7218:2007 þ Amd € ln, Germany: Beuth 1:2013); German version EN ISO 7218:2007 þ A1:2013. Ko Verlang GmbH. Anonymous. (2014b). Microbiology of the food chain - quantitative analysis of emetic €ln, Germany: toxin (cereulide) using LC-MS/MS (ISO 18465:2014-11-Draft). Ko Beuth Verlang GmbH. Banerjee, M., & Sarkar, P. K. (2003). Microbiological quality of some retail spices in India. Food Research International, 36(5), 469e474. Blanch, A. R., Mendez, J., Castel, S., & Reina, M. (2014). Comparison of procedures for the extraction of supernatants and cytotoxicity tests in Vero cells, applied to assess the toxigenic potential of Bacillus spp. and Lactobacillus spp., intended for use as probiotic strains. Journal of Microbiological Methods, 103, 64e69. Cardazzo, B., Negrisolo, E., Carraro, L., Alberghini, L., Patarnello, T., & Giaccone, V. (2008). Multiple-locus sequence typing and analysis of toxin genes in Bacillus cereus food-borne isolates. Applied and Environmental Microbiology, 74(3), 850e860. Ceuppens, S., Rajkovic, A., Heyndrickx, M., Tsilia, V., van De Wiele, T., Boon, N., et al. (2011). Regulation of toxin production by Bacillus cereus and its food safety implications. Critical Reviews in Microbiology, 37(3), 188e213. Cui, Y. F., Liu, X. Y., Dietrich, R., Martlbauer, E., Cao, J., Ding, S. Y., et al. (2016). Characterization of Bacillus cereus isolates from local dairy farms in China. Fems Microbiology Letters, 363(12), 6. €ffentlichte mikrobiologische Richt- und Warnwerte zur Beurteilung DGHM. (2011). Vero von Lebensmitteln. http://www.dghm.org/m_275. Doll, V. M., Ehling-Schulz, M., & Vogelmann, R. (2013). Concerted action of Sphingomyelinase and non-hemolytic enterotoxin in pathogenic Bacillus cereus. Plos One, 8(4). EFSA. (2005). Opinion of the scientific panel on biological hazards on Bacillus cereus and other Bacillus spp in foodstuffs. EFSA Journal, 175, 1e48. EFSA. (2016). Panel on biological hazards (BIOHAZ). Scientific opinion on the risks for public health related to the presence of Bacillus cereus and other Bacillus spp. including Bacillus thuringiensis in foodstuffs. EFSA Journal, 14(7), 93. EFSA. (2013). Panel on biological hazards (BIOHAZ). Scientific opinion on the risk posed by pathogens in food of non-animal origin. Part 1 (outbreak data analysis and risk ranking of food/pathogen combinations). EFSA Journal, 11(1:3025), 138. EFSA. (2014). Panel on additives and products or substances used in animal feed (FEEDAP). Scientific opinion on the safety and efficacy of Toyocerin® (Bacillus toyonensis) as a feed additive for rabbits for fattening, chickens for fattening, weaned piglets, pigs for fattening, sows for reproduction, cattle for fattening, and calves for rearing. EFSA Journal, 12(7), 17. Ehling-Schulz, M., Svensson, B., Guinebretiere, M. H., Lindback, T., Andersson, M., Schulz, A., et al. (2005). Emetic toxin formation of Bacillus cereus is restricted to a single evolutionary lineage of closely related strains. Microbiology Sgm, 151, 183e197. Ehling-Schulz, M., Vukov, N., Schulz, A., Shaheen, R., Andersson, M., Martlbauer, E., et al. (2005). Identification and partial characterization of the nonribosomal peptide synthetase gene responsible for cereulide production in emetic Bacillus cereus. Applied and Environmental Microbiology, 71(1), 105e113. Fagerlund, A., Ween, A., Lund, T., Hardy, S. P., & Granum, P. E. (2004). Genetic and functional analysis of the cytK family of genes in Bacillus cereus. MicrobiologySgm, 150, 2689e2697. Frederiksen, K., Rosenquist, H., Jorgensen, K., & Wilcks, A. (2006). Occurrence of natural Bacillus thuringiensis contaminants and residues of Bacillus thuringiensis-based insecticides on fresh fruits and vegetables. Applied and Environmental Microbiology, 72(5), 3435e3440. Frentzel, H., Dinh Thanh, M., Krause, G., Appel, B., & Mader, A. (2016). Quantification and differentiation of Bacillus cereus group species in spices and herbs by realtime PCR. Food Control. http://dx.doi.org/10.1016/j.foodcont.2016.11.028. Guinebretiere, M. H., Auger, S., Galleron, N., Contzen, M., De Sarrau, B., Buyser, De,

Please cite this article in press as: Frentzel, H., et al., Phylogenetic and toxinogenic characteristics of Bacillus cereus group members isolated from spices and herbs, Food Control (2016), http://dx.doi.org/10.1016/j.foodcont.2016.12.022

H. Frentzel et al. / Food Control xxx (2016) 1e9 et al. (2013). Bacillus cytotoxicus sp nov is a novel thermotolerant species of the Bacillus cereus group occasionally associated with food poisoning. International Journal of Systematic and Evolutionary Microbiology, 63, 31e40. Guinebretiere, M. H., Broussolle, V., & Nguyen-The, C. (2002). Enterotoxigenic profiles of food-poisoning and food-borne Bacillus cereus strains. Journal of Clinical Microbiology, 40(8), 3053e3056. Guinebretiere, M. H., Fagerlund, A., Granum, P. E., & Nguyen-The, C. (2006). Rapid discrimination of cytK-1 and cytK-2 genes in Bacillus cereus strains by a novel duplex PCR system. Fems Microbiology Letters, 259(1), 74e80. Guinebretiere, M. H., Velge, P., Couvert, O., Carlin, F., Debuyser, M. L., & NguyenThe, C. (2010). Ability of Bacillus cereus group strains to cause food poisoning varies according to phylogenetic affiliation (groups I to VII) rather than species affiliation. Journal of Clinical Microbiology, 48(9), 3388e3391. Haggblom, M. M., Apetroaie, C., Andersson, M. A., & Salkinoja-Salonen, M. S. (2002). Quantitative analysis of cereulide, the emetic toxin of Bacillus cereus, produced under various conditions. Applied and Environmental Microbiology, 68(5), 2479e2483. Hariram, U., & Labbe, R. (2015). Spore prevalence and toxigenicity of Bacillus cereus and Bacillus thuringiensis isolates from US retail spices. Journal of Food Protection, 78(3), 590e596. Hariram, U., & Labbe, R. G. (2016). Growth and inhibition by spices of growth from spores of enterotoxigenic Bacillus cereus in cooked rice. Food Control, 64, 60e64. Hoffmaster, A. R., Novak, R. T., Marston, C. K., Gee, J. E., Helsel, L., Pruckler, J. M., et al. (2008). Genetic diversity of clinical isolates of Bacillus cereus using multilocus sequence typing. Bmc Microbiology, 8. Jessberger, N., Krey, V. M., Rademacher, C., Bohm, M. E., Mohr, A. K., EhlingSchulz, M., et al. (2015). From genome to toxicity: A combinatory approach highlights the complexity of enterotoxin production in Bacillus cereus. Frontiers in Microbiology, 6, 15. Jimenez, G., Urdiain, M., Cifuentes, A., Lopez-Lopez, A., Blanch, A. R., Tamames, J., et al. (2013). Description of Bacillus toyonensis sp nov., a novel species of the Bacillus cereus group, and pairwise genome comparisons of the species of the group by means of ANI calculations. Systematic and Applied Microbiology, 36(6), 383e391. Jolley, K. A., & Maiden, M. C. J. (2010). BIGSdb: Scalable analysis of bacterial genome variation at the population level. BMC Bioinformatics, 11, 11. Jung, M. Y., Kim, J. S., Paek, W. K., Lim, J., Lee, H., Il Kim, P., et al. (2011). Bacillus manliponensis sp nov., a new member of the Bacillus cereus group isolated from foreshore tidal flat sediment. Journal of Microbiology, 49(6), 1027e1032. Jung, M. Y., Paek, W. K., Park, I. S., Han, J. R., Sin, Y., Paek, J., et al. (2010). Bacillus gaemokensis sp nov., isolated from foreshore tidal flat sediment from the yellow sea. Journal of Microbiology, 48(6), 867e871. Kim, H. J., Baek, S. Y., Lee, N., & Oh, S. W. (2013). Organic acid resistance and toxin gene profiles of B. Cereus isolated from red pepper powder. Journal of Food Safety, 33(3), 319e326. Kneifel, W., & Berger, E. (1994). Microbiological criteria of random samples of spices and herbs retailed on the Austrian market. Journal of Food Protection, 57(10), 893e901. Liu, Y., Lai, Q. L., Goker, M., Meier-Kolthoff, J. P., Wang, M., Sun, Y. M., et al. (2015). Genomic insights into the taxonomic status of the Bacillus cereus group. Scientific Reports, 5, 11. Liu, B., Liu, G. H., Hu, G. P., Cetin, S., Lin, N. Q., Tang, J. Y., et al. (2014). Bacillus bingmayongensis sp nov., isolated from the pit soil of emperor Qin's Terra-cotta

9

warriors in China. Antonie Van Leeuwenhoek International Journal of General and Molecular Microbiology, 105(3), 501e510. Marxen, S., Stark, T. D., Frenzel, E., Ruetschle, A., Luecking, G., Puerstinger, G., et al. (2015). Chemodiversity of cereulide, the emetic toxin of Bacillus cereus. Analytical and Bioanalytical Chemistry, 407(9), 2439e2453. Oliwa-Stasiak, K., Kolaj-Robin, O., & Adley, C. C. (2011). Development of real-time PCR assays for detection and quantification of Bacillus cereus group species: Differentiation of B. Weihenstephanensis and rhizoid B. Pseudomycoides isolates from milk. Applied and Environmental Microbiology, 77(1), 80e88. Priest, F. G., Barker, M., Baillie, L. W. J., Holmes, E. C., & Maiden, M. C. J. (2004). Population structure and evolution of the Bacillus cereus group. Journal of Bacteriology, 186(23), 7959e7970. Rivera, A. M. G., Granum, P. E., & Priest, F. G. (2000). Common occurrence of enterotoxin genes and enterotoxicity in Bacillus thuringiensis. Fems Microbiology Letters, 190(1), 151e155. Rosenquist, H., Smidt, L., Andersen, S. R., Jensen, G. B., & Wilcks, A. (2005). Occurrence and significance of Bacillus cereus and Bacillus thuringiensis in ready-toeat food. Fems Microbiology Letters, 250(1), 129e136. Sagoo, S. K., Little, C. L., Greenwood, M., Mithani, V., Grant, K. A., McLauchlin, J., et al. (2009). Assessment of the microbiological safety of dried spices and herbs from production and retail premises in the United Kingdom. Food Microbiology, 26(1), 39e43. Thorsen, L., Hansen, B. M., Nielsen, K. F., Hendriksen, N. B., Phipps, R. K., & Budde, B. B. (2006). Characterization of emetic Bacillus weihenstephanensis, a new cereulide-producing bacterium. Applied and Environmental Microbiology, 72(7), 5118e5121. Tourasse, N. J., Helgason, E., Klevan, A., Sylvestre, P., Moya, M., Haustant, M., et al. (2011). Extended and global phylogenetic view of the Bacillus cereus group population by combination of MLST, AFLP, and MLEE genotyping data. Food Microbiology, 28(2), 236e244. Trapecar, M., Leouffre, T., Faure, M., Jensen, H. E., Granum, P. E., Cencic, A., et al. (2011). The use of a porcine intestinal cell model system for evaluating the food safety risk of Bacillus cereus probiotics and the implications for assessing enterotoxigenicity. APMIS, 119(12), 877e884. Vassileva, M., Torii, K., Oshimoto, M., Okamoto, A., Agata, N., Yamada, K., et al. (2007). A new phylogenetic cluster of cereulide-producing Bacillus cereus strains. Journal of Clinical Microbiology, 45(4), 1274e1277. Wehrle, E., Didier, A., Moravek, M., Dietrich, R., & Martlbauer, E. (2010). Detection of Bacillus cereus with enteropathogenic potential by multiplex real-time PCR based on SYBR green I. Molecular and Cellular Probes, 24(3), 124e130. WHO. (2014). Ranking of low moisture foods in support of microbiological risk management. Preliminary report of FAO/WHO expert consultation on ranking of low moisture foods. Wielinga, P. R., Hamidjaja, R. A., Agren, J., Knutsson, R., Segerman, B., Fricker, M., et al. (2011). A multiplex real-time PCR for identifying and differentiating B. anthracis virulent types. International Journal of Food Microbiology, 145, S137eS144. Zahner, V., Silva, A., de Moraes, G. P., McIntosh, D., & de Filippis, I. (2013). Extended genetic analysis of Brazilian isolates of Bacillus cereus and Bacillus thuringiensis. Memorias Do Instituto Oswaldo Cruz, 108(1), 65e72. Zhou, G. P. Y., Liu, H. Z., He, J., Yuan, Y. M., & Yuan, Z. M. (2008). The occurrence of Bacillus cereus, B. thuringiensis and B. mycoides in Chinese pasteurized full fat milk. International Journal of Food Microbiology, 121(2), 195e200.

Please cite this article in press as: Frentzel, H., et al., Phylogenetic and toxinogenic characteristics of Bacillus cereus group members isolated from spices and herbs, Food Control (2016), http://dx.doi.org/10.1016/j.foodcont.2016.12.022