International Journal of Systematic and Evolutionary Microbiology (2013), 63, 31–40 DOI 10.1099/ijs.0.

030627-0

Bacillus cytotoxicus sp. nov. is a novel

thermotolerant species of the Bacillus cereus
Group occasionally associated with food poisoning
Marie-Hélène Guinebretière,1,2 Sandrine Auger,3,4 Nathalie Galleron,3,4
Matthias Contzen,5 Benoit De Sarrau,1,2 Marie-Laure De Buyser,6
Gilles Lamberet,3,4 Annette Fagerlund,7,8 Per Einar Granum,7
Didier Lereclus,3,4 Paul De Vos,9 Christophe Nguyen-The1,2
and Alexei Sorokin3,4
Correspondence 1
INRA, UMR408 Sécurité et Qualité des produits d’Origine Végétale, F-84000 Avignon, France
Marie-Hélène Guinebretière 2
Université d’Avignon, UMR408 Sécurité et Qualité des produits d’Origine Végétale,
marie-helene.guinebretiere@
F-84000 Avignon, France
avignon.inra.fr
3
INRA, UMR1319 MICALIS, F-78352 Jouy-en-Josas, France
4
AgroParisTech, UMR1319 MICALIS, F-78352 Jouy-en-Josas, France
5
Chemisches und Veterinäruntersuchungsamt Stuttgart, Schaflandstrasse 3/2, 70736 Fellbach,
Germany
6
ANSE, LERQUAP, Unité Caractérisation et Epidémiologie Bactérienne, F-94706 Maisons-Alfort
cedex, France
7
Department of Food Safety and Infection Biology, Norwegian School of Veterinary Science,
PO Box 8146, N-0033 Oslo, Norway
8
Department of Chemistry, Biotechnology and Food Science, Norwegian University of Life
Sciences, PO Box 5003, N-1432 Aas, Norway
9
Laboratory for Microbiology, Department Biochemistry and Microbiology, Ghent University,
K.L. Ledeganckstraat 35, B-9000 Gent, Belgium

An aerobic endospore-forming bacillus (NVH 391-98T) was isolated during a severe food
poisoning outbreak in France in 1998, and four other similar strains have since been isolated, also
mostly from food poisoning cases. Based on 16S rRNA gene sequence similarity, these strains
were shown to belong to the Bacillus cereus Group (over 97 % similarity with the current Group
species) and phylogenetic distance from other validly described species of the genus Bacillus
was less than 95 %. Based on 16S rRNA gene sequence similarity and MLST data, these novel
strains were shown to form a robust and well-separated cluster in the B. cereus Group, and
constituted the most distant cluster from species of this Group. Major fatty acids (iso-C15 : 0,
C16 : 0, iso-C17 : 0, anteiso-C15 : 0, iso-C16 : 0, iso-C13 : 0) supported the affiliation of these strains to
the genus Bacillus, and more specifically to the B. cereus Group. NVH 391-98T taxon was more
specifically characterized by an abundance of iso-C15 : 0 and low amounts of iso-C13 : 0 compared
with other members of the B. cereus Group. Genome similarity together with DNA–DNA
hybridization values and physiological and biochemical tests made it possible to genotypically and
phenotypically differentiate NVH 391-98T taxon from the six current B. cereus Group species.
NVH 391-98T therefore represents a novel species, for which the name Bacillus cytotoxicus sp.
nov. is proposed, with the type strain NVH 391-98T (5DSM 22905T5CIP 110041T).

Abbreviations: DDH, DNA–DNA hybridization; MLST, multilocus sequence typing.

The GenBank/EMBL/DDBJ accession numbers for the 16S rRNA gene of strains are described in Tables 1 and S1. The sequences of the MLST genes
can be found at http://mlstoslo.uio.no/index.html.
A supplementary figure and three supplementary tables are available with the online version of this paper.

030627 G 2013 IUMS Printed in Great Britain 31

M.-H. Guinebretière and others

INTRODUCTION cases. Together, they displayed phenotypic and genetic

differences to other B. cereus Group strains (Afchain et al.,
Bacillus species comprise psychrophilic to thermophilic
2008; Auger et al., 2008; Fagerlund et al., 2007; Rau et al.,
organisms, which allows them to colonize a wide range of
2009) and a particular thermotolerant ecotype known as
environments. Like other species of Bacillus, the Bacillus
phylogenetic group VII (Guinebretière et al., 2008). Here, it is
cereus Group species are metabolically diverse which enables
shown that these strains are closely related and represent a
them to develop in various environmental conditions. This
phylogenetically distinct group at present comprises six novel species of the B. cereus Group, for which the name
closely related species B. pseudomycoides (Nakamura, 1998), Bacillus cytotoxicus sp. nov. is proposed.
B. mycoides (Lechner et al., 1998; Nakamura, 1998; Nakamura
& Jackson, 1995), B. weihenstephanensis (Lechner et al., 1998),
B. cereus (Smith et al., 1952; Somerville & Jones, 1972), B. METHODS
thuringiensis (Nakamura, 1994; Smith et al., 1952; Somerville Strains and DNA sequences. NVH 391-98T (5DSM 22905T5CIP
& Jones, 1972) and B. anthracis (Logan et al., 1985; Smith 110041T) and close relative strains are listed in Table 1. Type strains
et al., 1952; Somerville & Jones, 1972). Adaptation to the used in this work (B. pseudomycoides DSM 12442T, B. mycoides CIP
environment has played a major role in shaping the evolution 103472T, B. weihenstephanensis WSBC 10204T, B. cereus ATCC
14579T, B. thuringiensis CIP 53.137T and B. subtilis CIP 52.65T) were
of these organisms, triggering a shift in growth temperature
all obtained from international collections, except strain WSBC
range for some species and emergence of specific ecotypes 10204T which was provided by the University of Munich. All strains
(Guinebretière et al., 2008). For example, B. mycoides and B. were maintained at 280 uC in micro-tubes containing 40 % glycerol
weihenstephanensis mainly derived from the natural envir- and all cultivated well on trypticase soy agar (30 g trypticase soy broth
onment (soils, rivers and plants) are preferentially associated and 15 g bacto agar; l21) and Luria agar (25 g Luria–Bertani broth
with cold thermal niches compared with other species and 15 g Bacto agar; l21) at 30–37 uC.
(Guinebretière et al., 2008; Lechner et al., 1998). The six Genomic sequences were obtained from Integrated Microbial
species are generally recognized as free-living soil organisms Genomes (IMG) (Markowitz et al., 2010) for all type strains (finished
but some of them are also potential pathogens of mammals or draft sequences), except for the B. weihenstephanensis type strain
and insects. Due to their broad ability to adaptat, B. cereus whose sequence was substituted for a genomic sequence of closely related
Group strains are readily recovered in foods, leading to food strain B. weihenstephanensis KBAB4. They included complete genomic
poisoning when conditions are favourable. It was during a sequence of NVH 391-98T, B. cereus ATCC 14579T, B. thuringiensis
ATCC 10792T, B. anthracis A0488T, B. anthracis Ames, B. mycoides DSM
food poisoning outbreak in France in 1998 that the singular
2048T, B. pseudomycoides DSM 12442T, B. weihenstephanensis KBAB4, B.
strain NVH 391-98T was first isolated and initially allocated to subtilis NCIB 3610T.
B. cereus. This event, resulting in three fatalities, represented
the most severe known case of diarrhoeal food poisoning in The entire 16S rRNA gene sequences from type strains of the genus
Bacillus were obtained from NCBI databases (refseq_rna and
France caused by a presumed B. cereus Group strain. The
refseq_genomic) when available, otherwise from the IMG genome
strain was intensively studied, and an important diarrhoeic database from which a consensus sequence from all 16 rRNA gene
enterotoxin, called cytotoxin K, was identified (Lund et al., copies was generated and used for analysis. For the recent isolates
2000) and further characterized (Brillard & Lereclus, 2004; 08CEB44BAC and CVUAS2833, 16S rRNA genes were amplified by
Fagerlund et al., 2004). Since then, four similar isolates have PCR and sequenced as previously described (Guinebretière et al.,
been reported, three of which were also linked to food 2008).
poisoning outbreaks. Two such cases were detected in France The concatenated sequences integrated into MLST analysis were those
and one in Germany. The strains were isolated from vegetable of genes described in the Tourasse-Helgason MLST scheme (Tourasse
purees and semolina incriminated in the food poisoning et al., 2006) and were obtained from http://mlstoslo.uio.no/index.html.

Table 1. Bacillus cytotoxicus strains included in this work

Strain Origin and year of isolation Other code/name Initial references 16S rRNA GenBank
accession no.
T
NVH 391-98 Vegetable puree, France, 1998, B. cereus 391-98, B. cereus Lund et al. (2000) AM747234
related to food poisoning ssp. cytotoxis, NVH 391-98
INRA AF2 Potato purée, France, 2003, B. cereus INRA AF2 Guinebretière et al. (2006), AM747232
related to food poisoning Fagerlund et al. (2007)
NVH 883/00 Spices, Norway, 2000 B. cereus NVH883/00 Guinebretière et al. (2006), AM747233
Fagerlund et al. (2007)
AFSSA 08CEB44 BAC Cooked Semolina, France, 2008, This paper JN790693
related to food poisoning
CVUAS2833 Potato purée, Germany, 2007, Rau et al. (2009) JN790694
related to food poisoning

32 International Journal of Systematic and Evolutionary Microbiology 63

 Bacillus cytotoxicus sp. nov.

For the recent isolates 08CEB44BAC and CVUAS2833, sequences were 2 uC higher than in the MIDI standard because it was found to be
obtained as previously described (Tourasse et al., 2006). close to the optimal temperature for all seven B. cereus Group species
(see Table 3 for temperature growth ranges). The effect of
Phylogenetic analysis. BLASTN software (http://blast.ncbi.nlm.nih. physiological age was minimized by harvesting cells from a streak
gov/Blast.cgi) was used to find the most closely related sequences plate on the overlap between the second and third quadrant on the
from the NCBI’s RNA database (Refseq_rna). A preliminary broad plate. After preparation, FAMEs were analysed by using gas
16S rRNA gene phylogeny including 103 species was generated (as chromatography-mass spectrometry (GC-MS) (Shimadzu QP2010
described in the following paragraph but running approximate system), as previously described (Brillard et al., 2010).
likelihood ratio tests instead of bootstraps for faster processing) and
The cell-wall diamino acid was determined from whole-cell hydro-
used to select 29 representative strains from the major robust clusters
lysat (4 M HCl, 100 uC, 16 h) subjected to thin-layer chromato-
(accession numbers can be found in Table S1, available in IJSEM
graphy on cellulose plates using the solvent system of Rhuland et al.
Online). The five B. cytotoxicus sequences (accession number found
(1955).
in Table 1), four B. cereus Group sequences (B. weihenstephanensis
AM747230T, B. mycoides AM747229T, B. cereus AF176322T, B. pseudo-
Phenotypic characterization. The strains were characterized
mycoides AM747226T), and consensus sequences from B. weihenstep-
phenotypically using API strips according to the method of Logan
hanensis KBAB4, B. anthracis A0488T and B. anthracis Ames genomes
& Berkeley (1984). Vegetative cells and sporangia were observed by
were added to the 29 representatives to the final phylogeny.
phase-contrast microscopy for rod shape, spore shape, spore position,
The 16S rRNA gene sequences were aligned using MUSCLE (Edgar, swollen sporangia and parasporal crystal. Electronic microscopy
2004). Phylogeny was reconstructed using PhyML 3.0 (Guindon et al., analysis was done by using the Technical platform MIMA2 (INRA,
2010). Maximum-likelihood searches were run under the GTR Jouy-en-Josas, France). Catalase production, starch hydrolysis,
(generalized time reversible) evolutionary model estimating all the Voges–Proskauer reaction, egg yolk reaction, anaerobic growth,
GTR model parameters and starting from a tree obtained using the reduction of nitrates, minimum and maximum growth temperatures
modified neighbour-joining algorithm BIONJ (Gascuel, 1997). Node all followed the standard tests given in Bergey’s Manual (Claus &
support was measured with 1000 bootstrap replicates. This same Berkeley, 1986; Logan & De Vos, 2009). For anaerobic growth,
procedure was applied to analyse the concatenated genes of the MLST aerobically incubated tubes were used as controls. Motility was tested
scheme. on Luria agar with 0.25 % agar, as described previously (Houry et al.,
2010). The tryptophan auxotrophy of NVH 391-98T and close
G+C content and DNA–DNA hybridizations. The G+C content of relatives was checked by observing the presence or absence of growth
DNA was determined by using HPLC (Mesbah et al., 1989) using the on Minimum Medium MOD (Duport et al., 2004; Rosenfeld et al.,
further specifications given by Logan et al. (2000). DNA–DNA 2005) supplemented with 5 g glucose l21 and with or without
hybridization (DDH) was performed using a modification of the tryptophan (1 g l21), compared to growth on a rich medium (TSA)
microplate method of Ezaki et al. (1989), as described by Willems and to growth of B. cereus ATCC 14579T as controls.
et al. (2001). A hybridization temperature of 37 uC (calculated with
correction for the presence of 50 % formamide) was used.
Predicted DDH value was calculated for all possible pairs of type RESULTS AND DISCUSSION
strains in the B. cereus Group (21 pairs) using a regression analysis
including experimental DDH values and genome similarity values: Genomic delimitation
predicted DDH51.066 (Index SIM) – 1.9689 (Fig. S1). The similarity
index between two genomes (Index SIM) was calculated by The 16S rRNA gene sequence of strain NVH 391-98T and
multiplying the ANI value (Goris et al., 2007) against the conserved of close relatives was a continuous stretch of 1532 bp. The
DNA % of the genome (Goris et al., 2007): (ANI%6conserved closest relatives of NVH 391-98T were the six species of the
DNA%)threshold 85 % /100 (Fig. S1 and Table S2). This index has the B. cereus group (97–98 % identity, Table S1 and Fig. 1).
advantage that it takes into account both the mean similarity of
conserved regions (represented by ANI) and the size of the conserved
Lower sequence similarities (¡95 % identity) were found
genome DNA, which is suitable to compare strains that significantly with all other validly described Bacillus species (Table S1).
differ in terms of size of conserved DNA, such as species of the B. The phylogenetic analysis indicated that strain NVH 391-
cereus Group. Both parameters were calculated using the ANI script 98T and close relative strains (INRA AF2, CVUAS2833,
(kindly provided by K. T. Konstantinidis, Michigan State University). 08CEB44BAC and NVH 883-00) formed a homogeneous
The final data are means of reciprocal values for each pair of strains. and robust cluster (Fig. 1) and that this cluster was easily
differentiated from all B. cereus Group species, as supported
Detection of the cytK-1 form of cytotoxin K. A pronounced
polymorphism for cytK gene has been previously observed, resulting by high bootstrap values. In contrast, 16S rRNA gene
in a specific form of the gene for NVH 391-98T (Fagerlund et al., sequence analysis was not able to clearly differentiate species
2004). As this form has been shown to be present in NVH 391-98T, of the B. cereus Group, except B. pseudomycoides. It is well
INRA AF2 and NVH883-00 (Fagerlund et al., 2007) but absent from known that these six species are genetically very close
all other species of the B. cereus Group among 420 tested strains (Turnbull et al., 2002). What is important here is that all
(Guinebretière et al., 2010), we used PCR to screen for the presence of these six species of the B. cereus Group are genetically
this cytK-1 form in the two other closely related strains, as described
previously (Guinebretière et al., 2006).
distinct from the NVH 391-98T cluster based on 16S rRNA
gene sequence analysis.
Chemotaxonomic characterization. Extraction of Fatty Acid MLST data allowed a more discriminatory analysis in the
Methyl Esters (FAMEs) was performed according to the standardized
MIDI protocol (http://www.microbialid.com/PDF/TechNote_101.
B. cereus Group (Fig. 2), supporting the previous data and
pdf). In our experiment, bacterial cells were obtained from culturing also highlighting NVH 391-98T cluster as a robust genetic
at 30 uC on trypticase soy broth agar (TSBA, 30 g trypticase soy broth entity and as the most distant cluster from other species in the
and 15 g Bacto agar; l21) for 24 h. The growth temperature used was B. cereus Group, followed by B. pseudomycoides. Sequences of

http://ijs.sgmjournals.org 33
M.-H. Guinebretière and others

B. weihenstephanensis WSBC 10204T

100 B. mycoides CIP 103472T
0.05 73 B. weihenstephanensis KBAB4
65 B. thuringiensis IAM12077
T
96 B. cereus ATCC 14579
B. anthracis A0488
100 95
B. anthracis Ames

99 B. pseudomycoides DSM 12442T

AFSSA 08CEB44 BAC

CVUAS2833
95
86 NVH 391-98T
T
100 NVH 883/00
INRA AF2
30
B. panaciterrae Gsoil 1517T
94 B. funiculus NAF001T
69
B. megaterium IAM 13418T
B. cohnii DSM 6307T
B. acidiceler CBD 119T
45
100 B. luciferensis LMG 18422T
17
57 B. amyloliquefaciens NBRC 15535
24 B. subtilis subsp. subtilis DSM 10T
B. atrophaeus JCM 9070T
99
14 B. pumilus ATCC 7061T
B. licheniformis ATCC 14580T
32
B. humi LMG 22167T
50
47 B. indicus Sd/3T
30 B. hemicellulosilyticus C-11T
36 B. pseudalcaliphilus DSM 8725T
22 B. halodurans DSM 497T
48
B. pseudofirmus DSM 8715T
B. clausii DSM 8716T
97
B. algicola F12
46 B. gelatini LMG 21880T
B. methanolicus NCIMB 13113T
81 B. fortis R-6514T
30 B. ginsengi ge14T
94 B. fumarioli LMG 17489T
37
B. novalis LMG 21837T
82 B. jeotgali YKJ-10T
79 B. foraminis CV53T
Geobacillus stearothermophilus IFO 12550T

Fig. 1. Maximum-likelihood phylogenetic tree based on 16S rRNA gene sequences of all strains under study. The phylogenetic
position of B. cytotoxicus (indicated in bold type) is shown among type strains of a selection of representatives in the genus
Bacillus. Bootstrap values are given at each branch point. Geobacillus stearothermophilus was used as an outgroup to root the
tree. The accession numbers of the strains used to generate the tree are given in Table S1. Bar, 0.05 substitutions per site.

34 International Journal of Systematic and Evolutionary Microbiology 63

 Bacillus cytotoxicus sp. nov.

the seven genes selected for one of the proposed MLST but this is congruent with previous DDH data. Except for
schemes (Tourasse et al., 2006) appeared to be identical for B. pseudomycoides, the distinctiveness of these species
the strains NVH 391-98T, INRA AF2 and CVUAS2833. seems to not always be strictly based on formal genomic
Sequencing types for the two other strains (08CEB44BAC and delimitation referring to DNA–DNA reassociation values
NVH883-00) each contained only two alleles that are different previously obtained (Lechner et al., 1998; Nakamura, 1994;
(not shown). Therefore, all five strains can be considered as Somerville & Jones, 1972). This creates difficulties in
members of a single clonal complex. Since the strains were genetically distinguishing B. thuringiensis strains from B.
isolated from very different locations, this indicates that the cereus strains (Guinebretière et al., 2008; Helgason et al.,
‘B. cytotoxicus’ species represents a fairly narrow lineage of 2000; Hill et al., 2004) and B. weihenstephanensis strains
closely related strains. from B. mycoides strains (Guinebretière et al., 2008). In any
case, NVH 391-98T cluster is clearly distinct based on
DNA–DNA hybridization (DDH) values were experiment-
DDH values, as well as is B. pseudomycoides.
ally measured for 11 pairs of type strains in the B. cereus
Group (mean of reciprocal values) and extensively The DNA G+C content of NVH 391-98T is 35.8 mol%,
predicted for all possible pairs of type strains (21 pairs) similar to the other B. cereus Group species.
using a regression analysis including experimental DDH
values and genome similarities of type strains (Table 2).
Strain NVH 391-98T showed very low DDH similarity to Phenotypic features
the type strains B. subtilis NCIB 3610T (,1 %) and low The fatty acid profiles of strains belonging to the NVH 391-
DDH similarity to the type strains of B. cereus ATCC 98T cluster (available as Table S3) were highly similar,
14579T (14.4 %), B. thuringiensis ATCC 10792T (14.4 %), B. composed of iso-C15 : 0 (36 %), C16 : 0 (11 %), anteiso-C15 : 0
anthracis A0488T (13.4 %), B. weihenstephanensis WSBC (11 %), iso-C17 : 0 (8 %), iso-C16 : 0 (7 %), iso-C13 : 0 (7 %),
10204T (13 to 15.4 %), B. mycoides DSM 2048T (14.7 %) iso-C14 : 0 (5 %), C16 : 1v-6 (3 %), anteiso-C17 : 0 (3 %), C14 : 0
and B. pseudomycoides DSM 12442T (20.5 %). These results (2 %), anteiso-C13 : 0 (2 %), iso-C16 : 1v-5 (1 %), and others
are congruent with 16S rRNA gene analysis and MLST that were found at ,1 %. Most of the major fatty acids (iso-
data, delimiting a new genomic species at a DDH level of C15 : 0, anteiso-C15 : 0, iso-C16 : 0, C16 : 0, iso-C17 : 0, anteiso-
around 13–20 % with the six other species of the B. cereus C17 : 0) were those typically found in the Bacillus genus
Group. However, with the B. cereus Group (as with other (Kämpfer, 1994; Song et al., 2000). Lower amounts of
Bacillus species), we faced a situation where the 16S rRNA anteiso-C15 : 0 and the presence of some particular fatty acids
gene sequence similarities used for species delimitation (isoC12 : 0, iso-C13 : 0, anteiso-C13 : 0) are specific to the B.
(.97 %) did not correlate with expected DDH-based cereus Group (Song et al., 2000), supporting the affiliation of
species delimitation (should be ¢70 %), as the 16S rRNA NVH 391-98T cluster to the B. cereus Group. In the B. cereus
gene seems to have evolved much more slowly than other Group, there were specific between-species differences in the
genes in this bacterial group. Some species of the B. cereus amounts of certain major fatty acids, particularly for B.
Group are not clearly distinguished based on DDH values, pseudomycoides DSM 12442T and NVH 391-98T cluster;

100 B. cereus ATCC14579T

80 B. thuringiensis ATCC10792T
100 B. anthracis Ames
B. weihenstephanensis WSBC 10204T
100
100 B. mycoides ATCC 6462T
B. pseudomycoides DSM 12442T
AFSSA 08CEB44 BAC
NVH 883/00
100
CVUAS2833
0.02 79 INRA AF2

NVH 391-98T

Fig. 2. Maximum-likelihood phylogenetic tree based on concatenated sequences from genes included in MLST scheme of
Tourasse et al. (2006). The phylogenetic position of B. cytotoxicus strains (indicated in bold type) is shown among type strains
of B. cereus Group species. Bootstrap values above 75 % are given at each branch point. Bar, 0.02 substitutions per site.

http://ijs.sgmjournals.org 35
M.-H. Guinebretière and others

NVH 391-98T cluster presented significantly higher amounts

100
nt
7
of iso-C15 : 0 and anteiso-C15 : 0 (see Table S3) and signifi-
cantly lower amounts of iso-C13 : 0 compared to other B.
Experimental DDH and G+C values from this study
cereus Group species. These characteristics are thus specific

100
nt
nt
6
to the NVH 391-98T cluster in our experimental conditions.
As for members of B. cereus Group and for most species in

100
80
60
nt
5
the genus Bacillus, the strain NVH391-98T possessed a cell-
wall type based on meso-diaminopimelic acid as the

100
59
58
nt
nt
4

diagnostic diamino acid, which is consistent with assign-

ment of the strain to the genus Bacillus. Considering the
Table 2. DNA–DNA hybridization (DDH) values and G+C mol% among representatives of B. cereus Group and the type species B. subtilis

phylogenetic position of strain NVH 391-98T, it shows the

100
90
59
52
nt
nt
3

peptidoglycan type A1c.

Strains of the NVH 391-98T cluster were all endospore-
100

21
nt
nt

nt
nt
nt
2

forming, rod-shaped, motile and catalase-positive bacteria,

which confirm their affiliation to the genus Bacillus. On
100

20

13

16
nt

nt
nt
nt API20E and API50CH, they demonstrated the major
1

specificities of the B. cereus Group (Table 3). In the B.

G+C content

cereus Group, strains of the NVH 391-98T cluster could be

(mol%)

differentiated from most of the other species by assimila-

35.9

35.8

35.7
35.2
35.8
34.7
nt
nt

tion of D-mannose, absence of starch hydrolysis, absence of

assimilation of sucrose, starch and glycogen, and a weak VP
reaction. Two highly specific features of the NVH 391-98T
Predicted DDH values from genome similarity at 85 % identity threshold and

100
8

taxon were the absence of trehalose assimilation and

absence of growth without tryptophan. The absence of
trehalose assimilation may be explained by the absence of
100
,1
7

genes coding for trehalose-6-phosphate hydrolase and the

trehalose-specific PTS system, as observed in the genome of
,1

NVH 391-98T. Concerning the absence of growth without

100
59
6

tryptophan, we observed that the genome of NVH 391-98T

does not contain the entire tryptophan biosynthesis operon
,1
100
86
68
5
G+C value from IMG

(trp), which differentiates it from all other available

genome sequences of the B. cereus Group (85 in total).
,1
100

Therefore, this strain and its close relatives were dependent

59
59
54
4

on tryptophan for growth on minimal media.

,1

Minimum and maximum growth temperatures tested

100
77
56
56
53
3

according to the standard test in Bergey’s Manual were

unable to distinguish most of the known species in the B.
,1
100

17
17
15
15
13
2

cereus Group (Table 3). In particular, the large growth

temperature ranges observed for B. cereus and B. thuringiensis
,1

are indicative of broad-ranging diversity and may be

100

21

15
15
14
14
13
1

explained by the fact that they each belong to several

G+C content

ecotypes (Guinebretière et al., 2008). However, strains of the

NVH 391-98T cluster showed specificity in that they grow
(mol%)

35.9

35.4

35.5
35.2
35.3
34.8
35.2
43.5

between 20 and 50 uC whereas strains of the other species

grew between 5–15 uC and 35–45 uC. A shift in growth
temperature range makes the NVH 391-98T cluster the only
3. B. weihenstephanensis KBAB4
1. B. cytotoxicus sp. nov. NVH

thermotolerant taxon in the B. cereus Group.

4. B. mycoides DSM 2048T
5. B. cereus ATCC 14579T
6. B. thuringiensis 10792T

8. B. subtilis NCBI 3610T

7. B. anthracis A0488T

Other typical characteristics

2. B. pseudomycoides

As for many B. cereus Group strains, spores of NVH 391-98T

are characteristically surrounded by the exosporium struc-
DSM12442T
nt, Not tested.

ture (Fig. 3). The exosporium, being relatively hydrophobic,

391-98T

is believed to be responsible for the high ability of B. cereus

Strain

Group bacteria to attach to inert surfaces (Faille et al., 2007;

Tauveron et al., 2006). This is one of the properties of these

36 International Journal of Systematic and Evolutionary Microbiology 63

 Bacillus cytotoxicus sp. nov.

Table 3. Characteristics of B. cereus Group type strains, and of other strains of the same species in bracket (data from this study and
from Logan & Berkeley, 1984; Logan et al., 1985)
Strains: 1, B. cytotoxicus NVH 391-98T (n55 strains); 2, B. pseudomycoides DSM 12442T (n57 strains); 3, B. cereus ATCC 14579T (n511 strains) ; 4,
B. thuringiensis CIP 53137T (n58 strains); 5, B. weihenstephanensis WSBC 10204T (n55 strains); 6, B. mycoides CIP 102472T (n53 strains); 7, B.
anthracis species (data from Logan et al., 1985). Symbols: +, positive for the type strain and 90 % or more of strains; 2, negative for the type strain
and for 90 % or more of strains; w, weakly positive; d, 11–89 % of positive B. anthracis strains. Numbers in round or square brackets indicate the
percentage of positive strains recorded in the species when this character appears variable with strains of this study (square brackets) or with strains
from Logan & Berkeley (1984) (round brackets and italics). All strains gave positive results in API50CH for D-glucose, D-fructose, aesculin, maltose,
and negative results for erythritol, D-arabinose, L-arabinose, D-xylose, L-xylose, D-adonitol, b-methylxyloside, L-sorbose, L-rhamnose, dulcitol,
inositol, D-manitol, D-sorbitol, a-methyl-D-mannoside, melibiose, inulin, melezitose, raffinose, xylitol, D-lyxose, D-tagatose, D-fucose, L-fucose, D-
arabitol, L-arabitol, 2-ketogluconate, 5-ketogluconate. All strains gave negative results in API20E tests for lysine decarboxylase, ornithine
decarboxylase, H2S production, urease, tryptophan desaminase and indole production.

Characteristics 1 2 3 4 5 6 7

Common general
characteristics
Cell diameter .1.0 mm + + + + + + +
Utilization of mannitol 2 2 2 2 2 2 2
Egg yolk lecithinase +w +w + + + + +
Anaerobic growth +w + + + + + +
Differential general
characteristics
Rhyzoidal colony - + 2 2 2 + 2
Parasporal crystal 2 2 2 + 2 2 2
Starch hydrolysis 2 2 + [67] + + + +
Growth without tryptophan 2 + + + + + +
Voges–Proskauer test +w + + + + [60] + [67] +
API 20E tests
VP +w + + + + 60] + [60] +
Nitrate reduction + + + [81] + + + +
ONPG 2 2 2 2 2 2 (32) 2
ADH + + [71] + [70] + [67] + [40] 2 (36) 2
Citrate (Simmons’) - [33] - [11] + [34] + [33] 2 2 (60) 2
Gelatin + + [86] + + + [80] + [67] d
API 50CH tests
Glycerol - +w [17w] 2 (90) 2 (92) +w [33w] 2 (96) 2
Ribose + [67] + + - [81] +w +w (76) +
Galactose - - - - [38] - - (32) 2
D-Mannose + - - [18] + [25] 2 [67] 2 2
a-Methyl-D-glucoside - - - - 2 [33] 2 (12) 2
N-Acetylglucosamine + + + + [69] + + +
Amygdalin +w [67] 2 +w [23w] - +w [83] - [50] 2
Arbutin + 2 [33] + [82] + [63] + + (84) d
Salicin + + [67] + [64] + [75] + + (80) 2
Cellobiose + 2 [33] + [55] + [63] +w [83] +w [75] 2
Sucrose 2 2 [33] + [64] + [38] 2 [33] + [50] +
Trehalose 2 + + + + + +
Starch 2 + + [73] + [80] + [67] + +
Glycogen 2 + + [73] + [88] + [67] + +
P-Gentiobiose 2 2 2 (9) 2 - [33] 2 2
Turanose 2 2 2 (16) 2 (11) 2 2 (24) 2
Gluconate 2 2 2 (36) 2 2 2 (12) 2
Minimum growth 20 uC 10 uC 10 uC 10 uC 5 uC 5 uC .10 uC
temperature (8–10 uC) (7–15 uC) (7–15 uC) (5–7 uC) (5–7 uC)
Maximum growth 50 uC 40 uC 45 uC 45 uC 37 uC 37 uC ,50 uC
temperature (40–43 uC) (40–45 uC) (40–45 uC) (35–37 uC) (35–37 uC)

http://ijs.sgmjournals.org 37
M.-H. Guinebretière and others

Fig. 3. Electron microscopy images of B.

cytotoxicus cells. Vegetative cells of NVH 391-
98T (a) and spores of NVH 391-98T (b) and
INRA AF2 (c) strains, respectively. The light
body, attached to the black spore on (b) and
(c), represents exosporium. Note the filaments
surrounding the vegetative cells of NVH 391-
98T (a). The scale bars correspond to 2 mm
(a) and 1 mm (b) and (c).

bacteria that allow them to persist on washable industrial difference in the chromosome sizes (4.1 compared to 5.2–
surfaces and thus spark food contamination issues and, in 5.4 Mb) (Lapidus et al., 2008).
the worst-case scenario, food poisoning.
All strains of the NVH 391-98T cluster carried the cytK Conclusion
gene under the cytK-1 form that corresponds to a
The very high genetic proximity of NVH 391-98T taxon to
pronounced polymorphism (Fagerlund et al., 2004) and
the six B. cereus Group species, and its similarity with their
is the higher cytotoxic form (Fagerlund et al., 2007). The
phenotypic features, prompts us to consider this taxon as
cytK-1 form is distinctive of the NVH 391-98T taxon, since
belonging to the B. cereus Group. In addition, MLST data
it has never been shown by tracking on 425 B. cereus Group
and DDH values provide ample evidence that this taxon
strains for any other species of the B. cereus Group
should be considered as a novel genomic species.
(Guinebretière et al., 2010). The resulting enterotoxin
Chemotaxomomic, biochemical and physiological data
(cytotoxin K-1) was first isolated for NVH 391-98T (Lund
correlate the genetic results, providing phenotypic differ-
et al., 2000), which is able to synthesize this toxin in high
entiation of this novel species from the six current B. cereus
amounts (Brillard & Lereclus, 2004; Fagerlund et al., 2007)
Group species.
and is considered particularly virulent since it has been
connected to three deaths. The CytK-1 form has been
exploited for a PCR diagnostic method that is able to detect Description of Bacillus cytotoxicus
B. cytotoxicus strains and has been validated on numerous
Bacillus cytotoxicus (cy.to.to9xi.cus. Gr. n. kutos, hollow,
B. cereus Group strains (Guinebretière et al., 2006, 2010).
hold of a ship; N.L. pref. cyto-, prefix denoting pertaining
However, not all strains are able to produce cytotoxin K
to a cell; N.L. adj. toxicus -a -um (from L. n. toxicum,
(Fagerlund et al., 2007). The PCR-based diagnostic method
poison), toxic; N.L. masc. adj. cytotoxicus, cytotoxic,
is therefore suitable for detecting B. cytotoxicus strains but
referring to cytotoxin K, an enterotoxin isolated and
does not give a level of cytotoxicity for the detected strains.
described from the type strain).
Among MLST genes, panC originated from a pilot
Isolated from rehydrated foods connected to food poison-
experiment on multiple locus sequencing of B. cereus
ing outbreaks, unknown natural niche. Gram-positive,
strains (Candelon et al., 2004). As the phylogeny of panC
facultative anaerobic (although aerobic growth is faster),
closely follows the global phylogeny of the B. cereus Group
motile rods (¢1.0 mm). Cell morphology similar to B.
(Guinebretière et al., 2008), it can be used as a tool for
cereus: cells occur singly, in pairs, occasionally in short chains
tracking strains of NVH 391-98 taxon (phylogenetic group
or filaments. Endospores are mainly ellipsoidal and lie central
VII) at https://www.tools.symprevius.org/Bcereus/english.
to subterminal position in non-swollen sporangia. When
php (Guinebretière et al., 2010). This is thus another rapid
grown for 24 h on TSA at 37 uC, colonies are 1 mm
method for identifying strains of B. cytotoxicus, through
diameter, cream-coloured, round, with a flat and shiny
their highly specific panC polymorphism.
surface; may become matt and with slightly irregular margins
The complete genomic sequence of the strain NVH 391- with age. Like other species of the B. cereus group, is egg-yolk
98T confirmed genetic divergence with other representa- lecithinase-positive and mannitol-negative, resulting in
tives of the B. cereus Group, especially highlighted by large typical pink colony with white precipitate around the colony

38 International Journal of Systematic and Evolutionary Microbiology 63

 Bacillus cytotoxicus sp. nov.

on Mossel medium, generally less pronounced than for B. Edgar, R. C. (2004). MUSCLE: multiple sequence alignment with high
cereus, or resulting in atypical precipitate under the colony. accuracy and high throughput. Nucleic Acids Res 32, 1792–1797.
Optimal growth occurs at 30–37 uC, the maximum growth Ezaki, T., Hashimoto, Y. & Yabuuchi, E. (1989). Fluorometric
temperature is 50 uC. Can be differentiated from other deoxyribonucleic acid deoxyribonucleic acid hybridization in micro-
species of the B. cereus Group by maximum growth at 50 uC dilution wells as an alternative to membrane-filter hybridization in
which radioisotopes are used to determine genetic relatedness among
and minimum growth at 20 uC, by absence of starch
bacterial strains. Int J Syst Bacteriol 39, 224–229.
hydrolysis, by absence of growth on synthetic media without
Fagerlund, A., Ween, O., Lund, T., Hardy, S. P. & Granum, P. E.
tryptophan, and weak VP reaction. API20E and API50CH
(2004). Genetic and functional analysis of the cytK family of genes in
are summarized in Table 3 highlighting the following specific Bacillus cereus. Microbiology 150, 2689–2697.
characteristics for B. cytotoxicus: VP weak, D-mannose-
Fagerlund, A., Brillard, J., Fürst, R., Guinebretière, M. H. & Granum,
positive, sucrose-negative, trehalose-negative, starch and P. E. (2007). Toxin production in a rare and genetically remote cluster
glycogen-negative. The major fatty acids are similar to those of strains of the Bacillus cereus group. BMC Microbiol 7, 43.
of other B. cereus Group species, with the particularity that
Faille, C., Tauveron, G., Le Gentil-Lelièvre, C. & Slomianny, C.
iso-C13 : 0 is in lower amounts and iso-C15 : 0 and anteiso- (2007). Occurrence of Bacillus cereus spores with a damaged
C15 : 0 are in higher amounts. The cell-wall peptidoglycan exosporium: consequences on the spore adhesion on surfaces of food
contains meso-diaminopimelic acid. processing lines. J Food Prot 70, 2346–2353.
The type strain is NVH 391-98T (5DSM 22905T5 Gascuel, O. (1997). BIONJ: an improved version of the NJ algorithm
CIP110041T) isolated during a severe food poisoning based on a simple model of sequence data. Mol Biol Evol 14, 685–695.
outbreak in France in 1998. The genomic DNA G+C Goris, J., Konstantinidis, K. T., Klappenbach, J. A., Coenye, T.,
content of the type strain is 35.8 mol%. Vandamme, P. & Tiedje, J. M. (2007). DNA-DNA hybridization
values and their relationship to whole-genome sequence similarities.
Int J Syst Evol Microbiol 57, 81–91.
ACKNOWLEDGEMENTS Guindon, S., Dufayard, J. F., Lefort, V., Anisimova, M., Hordijk, W. &
Gascuel, O. (2010). New algorithms and methods to estimate
We thank Christine Longin (MIMA2 technical platform, CRJ INRA, maximum-likelihood phylogenies: assessing the performance of
Jouy-en-Josas) for leading the electron microscopy analysis and K. T. PhyML 3.0. Syst Biol 59, 307–321.
Konstantinidis for providing the ANI script and coaching us on how Guinebretière, M. H., Fagerlund, A., Granum, P. E. & Nguyen-The, C.
best to use it. This publication made use of the University of Oslo’s (2006). Rapid discrimination of cytK-1 and cytK-2 genes in Bacillus
Bacillus cereus group Multilocus and MultiData Typing Website cereus strains by a novel duplex PCR system. FEMS Microbiol Lett 259,
(http://mlstoslo.uio.no). The work was partially supported by the 74–80.
French National Research Agency (project ANR-05-PNRA-013) and
by the INRA’s MICA department. Guinebretière, M. H., Thompson, F. L., Sorokin, A., Normand, P.,
Dawyndt, P., Ehling-Schulz, M., Svensson, B., Sanchis, V., Nguyen-
The, C. & other authors (2008). Ecological diversification in the
Bacillus cereus Group. Environ Microbiol 10, 851–865.
REFERENCES
Guinebretière, M. H., Velge, P., Couvert, O., Carlin, F., Debuyser,
Afchain, A. L., Carlin, F., Nguyen-The, C. & Albert, I. (2008). M. L. & Nguyen-The, C. (2010). Ability of Bacillus cereus group strains
Improving quantitative exposure assessment by considering genetic to cause food poisoning varies according to phylogenetic affiliation
diversity of B. cereus in cooked, pasteurised and chilled foods. Int J (groups I to VII) rather than species affiliation. J Clin Microbiol 48,
Food Microbiol 128, 165–173. 3388–3391.
Auger, S., Galleron, N., Bidnenko, E., Ehrlich, S. D., Lapidus, A. & Helgason, E., Okstad, O. A., Caugant, D. A., Johansen, H. A., Fouet, A.,
Sorokin, A. (2008). The genetically remote pathogenic strain Mock, M., Hegna, I. & Kolstø, A. B. (2000). Bacillus anthracis, Bacillus
NVH391-98 of the Bacillus cereus group is representative of a cluster cereus, and Bacillus thuringiensis–one species on the basis of genetic
of thermophilic strains. Appl Environ Microbiol 74, 1276–1280. evidence. Appl Environ Microbiol 66, 2627–2630.
Brillard, J. & Lereclus, D. (2004). Comparison of cytotoxin cytK Hill, K. K., Ticknor, L. O., Okinaka, R. T., Asay, M., Blair, H., Bliss, K. A.,
promoters from Bacillus cereus strain ATCC 14579 and from a B. Laker, M., Pardington, P. E., Richardson, A. P. & other authors
cereus food-poisoning strain. Microbiology 150, 2699–2705. (2004). Fluorescent amplified fragment length polymorphism analysis
of Bacillus anthracis, Bacillus cereus, and Bacillus thuringiensis isolates.
Brillard, J., Jéhanno, I., Dargaignaratz, C., Barbosa, I., Ginies, C.,
Appl Environ Microbiol 70, 1068–1080.
Carlin, F., Fedhila, S., Nguyen-the, C., Broussolle, V. & Sanchis, V.
(2010). Identification of Bacillus cereus genes specifically expressed during Houry, A., Briandet, R., Aymerich, S. & Gohar, M. (2010).
growth at low temperatures. Appl Environ Microbiol 76, 2562–2573. Involvement of motility and flagella in Bacillus cereus biofilm
Candelon, B., Guilloux, K., Ehrlich, S. D. & Sorokin, A. (2004). Two
formation. Microbiology 156, 1009–1018.
distinct types of rRNA operons in the Bacillus cereus group. Kämpfer, P. (1994). Limits and Possibilities of Total Fatty-Acid
Microbiology 150, 601–611. Analysis for Classification and Identification of Bacillus Species. Syst
Claus, D. & Berkeley, R. C. W. (1986). Genus Bacillus Cohn 1872, Appl Microbiol 17, 86–98.
174AL. In Bergey’s Manual Of Systematic Bacteriology, pp. 1105–1139. Lapidus, A., Goltsman, E., Auger, S., Galleron, N., Ségurens, B.,
Edited by P. H. A. Sneath, N. S. Mair, M. E. Sharpe & J. G. Holt. Dossat, C., Land, M. L., Broussolle, V., Brillard, J. & other authors
Baltimore, Md: Williams & Wilkins. (2008). Extending the Bacillus cereus group genomics to putative food-
Duport, C., Thomassin, S., Bourel, G. & Schmitt, P. (2004). borne pathogens of different toxicity. Chem Biol Interact 171, 236–249.
Anaerobiosis and low specific growth rates enhance hemolysin BL Lechner, S., Mayr, R., Francis, K. P., Prüss, B. M., Kaplan, T.,
production by Bacillus cereus F4430/73. Arch Microbiol 182, 90–95. Wiessner-Gunkel, E., Stewart, G. S. & Scherer, S. (1998). Bacillus

http://ijs.sgmjournals.org 39
M.-H. Guinebretière and others

weihenstephanensis sp. nov. is a new psychrotolerant species of the using cultural methods, LC-MS/MS, PCR, and infrared spectroscopy
Bacillus cereus group. Int J Syst Bacteriol 48, 1373–1382. in consideration of thermotolerant isolates]. Berl Munch Tierarztl
Logan, N. A. & Berkeley, R. C. W. (1984). Identification of Bacillus Wochenschr 122, 25–36.
strains using the API system. J Gen Microbiol 130, 1871–1882. Rhuland, L. E., Work, E., Denman, R. F. & Hoare, D. S. (1955). The
Logan, N. A. & De Vos, P. (2009). Genus Bacillus Cohn 1872. In behaviour of the isomers of a, e-diaminopimelic acid on paper
Bergey’s Manual of Systematic Bacteriology 2nd edn, pp. 21–128. Edited chromatograms. J Am Chem Soc 77, 4844–4846.
by P. De Vos & G. M. Garrity, D. N. R. Jones, W. Frieg, F. A. LudWig, Rosenfeld, E., Duport, C., Zigha, A. & Schmitt, P. (2005).
K.-H. Rainey, W. B. Schleifer & Whitman. New York: Springer Characterization of aerobic and anaerobic vegetative growth of the
Dordrecht Heidelberg London. food-borne pathogen Bacillus cereus F4430/73 strain. Can J Microbiol
Logan, N. A., Carman, J. A., Melling, J. & Berkeley, R. C. W. (1985). 51, 149–158.
Identification of Bacillus anthracis by API tests. J Med Microbiol 20, Smith, N. R., Gordon, R. E. & Clarck, F. E. (1952). Aerobic Spore-
75–85. forming Bacteria Monograph no 16. Wsahington, DC: US Department
Logan, N. A., Lebbe, L., Hoste, B., Goris, J., Forsyth, G., Heyndrickx, M., of Agriculture.
Murray, B. L., Syme, N., Wynn-Williams, D. D. & De Vos, P. (2000). Somerville, H. J. & Jones, M. L. (1972). DNA competition studies
Aerobic endospore-forming bacteria from geothermal environments in within the Bacillus cereus group of bacilli. J Gen Microbiol 73, 257–
northern Victoria Land, Antarctica, and Candlemas Island, South 265.
Sandwich archipelago, with the proposal of Bacillus fumarioli sp. nov.
Int J Syst Evol Microbiol 50, 1741–1753. Song, Y., Yang, R., Guo, Z., Zhang, M., Wang, X. & Zhou, F. (2000).
Distinctness of spore and vegetative cellular fatty acid profiles of some
Lund, T., De Buyser, M. L. & Granum, P. E. (2000). A new cytotoxin aerobic endospore-forming bacilli. J Microbiol Methods 39, 225–241.
from Bacillus cereus that may cause necrotic enteritis. Mol Microbiol
38, 254–261. Tauveron, G., Slomianny, C., Henry, C. & Faille, C. (2006). Variability
among Bacillus cereus strains in spore surface properties and influence
Markowitz, V. M., Chen, I. A., Palaniappan, K., Chu, K., Szeto, E., on their ability to contaminate food surface equipment. Int J Food
Grechkin, Y., Ratner, A., Anderson, I., Lykidis, A. & other authors (2010). Microbiol 110, 254–262.
The Integrated Microbial Genomes system: an expanding comparative
analysis resource. Nucleic Acids Res 38 (Database), D382–D390. Tourasse, N. J., Helgason, E., Økstad, O. A., Hegna, I. K. & Kolstø,
A. B. (2006). The Bacillus cereus group: novel aspects of population
Mesbah, M., Premachandran, U. & Whitman, W. B. (1989). Precise
structure and genome dynamics. J Appl Microbiol 101, 579–593.
Measurement of the G+C Content of Deoxyribonucleic-Acid by
High-Performance Liquid-Chromatography. Int J Syst Bacteriol 39, Turnbull, P. C. B., Jackson, B. J., Hill, K. K., Keim, P., Kolsto, A. B. &
159–167. Beecher, D. J. (2002). Longstanding taxonomic enigmas within the
‘Bacillus Cereus Group’ are on the verge of being resolved by far-
Nakamura, L. K. (1994). DNA relatedness among Bacillus thuringien-
reaching molecular developments: forecasts on the possible outcome
sis serovars. Int J Syst Bacteriol 44, 125–129.
by an ad hoc team. In Applications and Systematics of Bacillus and
Nakamura, L. K. (1998). Bacillus pseudomycoides sp. nov. Int J Syst Relatives, pp. 23–36. Edited by R. C. W. Berkeley, M. Heyndrickx,
Bacteriol 48, 1031–1035. N. A. Logan & P. De Vos. Oxford, UK: Blackwell Science.
Nakamura, L. K. & Jackson, M. A. (1995). Clarification of the Willems, A., Doignon-Bourcier, F., Goris, J., Coopman, R., de
taxonomy of Bacillus mycoides. Int J Syst Bacteriol 45, 46–49. Lajudie, P., De Vos, P. & Gillis, M. (2001). DNA-DNA hybridization
Rau, J., Perz, R., Klittich, G. & Contzen, M. (2009). [Cereulide forming study of Bradyrhizobium strains. Int J Syst Evol Microbiol 51, 1315–
presumptive Bacillus cereus strains from food–differentiating analyses 1322.

40 International Journal of Systematic and Evolutionary Microbiology 63

Supplementary data Table S1. Closely related 16S rDNA sequences in the genus Bacillus and in the
most representative and related genus found in NCBI database (refseq_rna and refseq_genomic).
Sequences selected for the final phylogenetic analysis are in boldface.

Total Covera- E- %
Accession Description
Score ge value Identity

NZ_CM000745.1 T
Bacillus pseudomycoides DSM 12442, genome 2704 100% 0.0 98%
NC_007530.2 Bacillus anthracis str. 'Ames Ancestor', genome 2687 100% 0,0 98%
NC_004722.1 T
Bacillus cereus ATCC 14579, genome 2687 100% 0,0 98%
T
NR 043403.1 Bacillus thuringiensis strain lAM 12077 2562 96% 0,0 97%
NZ_CM000742.1 T
Bacillus mycoides DSM 2048, genome 2654 100% 0.0 97%
T
NR_024697.1 Bacillus weihenstephanensis strain DSM 11821 2628 99% 0,0 97%
T
NR_036880.1 Bacillus mycoides strain 273 2595 97% 0,0 97%
T
NR_025373.1 Bacillus shackletonii strain LMG 18435 2418 97% 0,0 95%
T
NR_026138.1 Bacillus cohnii strain DSM 6307 2405 97% 0,0 95%
T
NR_025511.1 Bacillus luciferensis strain LMG 18422 2398 97% 0,0 95%
T
NR_028624.1 Bacillus funiculus strain NAF001 2446 99% 0,0 95%
T
NR_041942.1 Bacillus acidicola strain 105-2 2462 99% 0,0 95%
T
NR_025240.1 Bacillus marisflavi strain TF-11 2388 97% 0,0 95%
T
NR_041378.1 Bacillus ginsengihumi strain Gsoil 114 2370 96% 0,0 95%
T
NR_043084.1 Bacillus koreensis strain BR030 2158 88% 0,0 95%
T
NR_043774.1 Bacillus acidiceler strain CBD 119 2353 96% 0,0 95%
T
NR_024817.1 Bacillus asahii strain MA001 2361 97% 0,0 94%
T
NR_042286.1 Bacillus herbersteinensis strain D-1,5a 2296 94% 0,0 94%
T
NR_026144.1 Bacillus halmapalus strain DSM 8723 2350 97% 0,0 94%
T
NR_024691.1 Bacillus flexus strain IFO15715 2385 98% 0,0 94%
T
NR_043083.1 Bacillus seohaeanensis strain BH724 2154 89% 0,0 94%
T
NR_026010.1 Bacillus sporothermodurans strain M215 2326 96% 0,0 94%
T
NR_041377.1 Bacillus pocheonensis strain Gsoil 420 2320 96% 0,0 94%
T
NR_025241.1 Bacillus aquimaris strain TF-12 2324 97% 0,0 94%
T
NR_042274.1 Bacillus foraminis strain CV53 2370 99% 0,0 94%
T
NR_042136.1 Bacillus simplex strain DSM 1321 2348 98% 0,0 94%
T
NR_029002.1 Bacillus drentensis strain IDA1967 2213 92% 0,0 94%
T
NR_043334.1 Bacillus niabensis strain 4T19 2278 95% 0,0 94%
T
NR_043325.1 Bacillus oleronius strain ATCC 700005 2381 99% 0,0 94%
T
NR_041379.1 Bacillus panaciterrae strain Gsoil 1517 2281 96% 0,0 94%
T
NR_025060.1 Bacillus jeotgali strain YKJ-10 2311 97% 0,0 94%
T
NR_040792.1 Bacillus lentus strain NCIMB8773 2361 99% 0,0 94%
T
NR_025557.1 Bacillus aeolius strain 4-1 2156 90% 0,0 94%
T
NR_040852.1 Bacillus horikoshii strain DSM 8719 2314 97% 0,0 94%
T
NR_036766.1 Bacillus bataviensis strain IDA1115 2300 97% 0,0 94%
T
NR_043401.1 Bacillus megaterium strain IAM 13418 2272 96% 0,0 94%
T
NR_040985.1 Bacillus methanolicus strain NCIMB 13113 2298 97% 0,0 94%
T
NR_044546.1 Bacillus nealsonii strain DSM 15077 2307 97% 0,0 94%
T
NR_024689.1 Bacillus atrophaeus strain JCM9070 2313 97% 0,0 94%
 T
NR_042619.1 Bacillus isabeliae strain CVS-8 2331 99% 0,0 94%
T
NR_042339.1 Bacillus aerophilus strain28K 2337 99% 0,0 94%
T
NR_042336.1 Bacillus stratosphericus strain41KF2a 2337 99% 0,0 94%
T
NR_024695.1 Bacillus niacini strain IFO15566 2322 98% 0,0 94%
T
NC_006322.1 Bacillus licheniformis ATCC 14580 2359 100% 0.0 94%
T
NR_042337.1 Bacillus altitudinis strain41KF2b 2254 96% 0,0 94%
T
NR_042638.1 Bacillus halotolerans strain DSM 8802 2337 99% 0,0 94%
T
NR_026139.1 Bacillus pseudofirmus strain DSM 8715 2278 97% 0,0 94%
T
NR_042726.1 Bacillus circulans strain name not vailable 2268 97% 0,0 94%
T
NR_042168.1 Bacillus novalis strain LMG 21837 2274 97% 0,0 94%
T
NR_025626.1 Bacillus humi strain LMG 22167 2281 97% 0,0 94%
T
NR_027552.1 Bacillus subtilis subsp. subtilis strain DSM 10 2300 98% 0,0 94%
T
NR_040793.1 Bacillus psychrosaccharolyticus strain ATCC 23296 2283 97% 0,0 94%
T
NR_043268.1 Bacillus idriensis strain SMC 4352-2 2165 92% 0,0 94%
T
NR_024696.1 Bacillus vallismortis strain DSM11031 2318 99% 0,0 94%
T
NR_041641.1 Bacillus azotoformans strain NBRC 15712 2233 95% 0,0 94%
T
NR_024693.1 Bacillus mojavensis strain IFO15718 2311 98% 0,0 94%
T
NR_044538.1 Bacillus korlensis strain ZLC-26 2224 95% 0,0 94%
T
NR_043762.1 Bacillus thioparans strain BMP-1 2220 95% 0,0 94%
T
NR_024690.1 Bacillus carboniphilus strain JCM9731 2268 97% 0,0 93%
T
NR_025842.1 Bacillus firmus strain IAM 12464 2239 95% 0,0 93%
T
NR_041275.1 Bacillus boroniphilus strain T-15Z 2259 96% 0,0 93%
NR_044828.1 Bacillus benzoevoransT strain NCIMB 12555 2161 92% 0,0 93%
T
NR_043682.1 Bacillus kribbensis strain BT080 2246 96% 0,0 93%
T
NR_043015.1 Bacillus litoralis strain SW-211 2266 97% 0,0 93%
T
NR_041455.1 Bacillus amyloliquefaciens strain NBRC 15535 2217 95% 0,0 93%
T
NR_025264.1 Bacillus hwajinpoensis strain SW-72 2263 97% 0,0 93%
T
NR_044170.1 Bacillus butanolivorans strain K9 2204 94% 0,0 93%
T
NR_041794.1 Bacillus safensis strain FO-036b 2159 93% 0,0 93%
T
NR_043242.1 Bacillus pumilus strain ATCC 7061 2159 93% 0,0 93%
T
NR_029077.1 Bacillus algicola strain F12 2322 99% 0,0 93%
T
NR_036987.1 Bacillus smithii strain NRS-173 2289 98% 0,0 93%
T
NR_025591.1 Bacillus soli strain R-16300 2161 93% 0,0 93%
T
NR_043484.1 Bacillus okhensis strain Kh10-101 2200 94% 0,0 93%
T
NR_025590.1 Bacillus vireti strain R-15447 2141 92% 0,0 93%
T
NR_025122.1 Bacillus endophyticus strain 2DT 2204 95% 0,0 93%
T
NR_044204.1 Bacillus alkalinitrilicus strain ANL-iso4 2193 95% 0,0 93%
T
NR_028620.1 Bacillus akibai strain 1139 2248 97% 0,0 93%
T
NR_025370.1 Bacillus fumarioli strain LMG 17489 2187 94% 0,0 93%
T
NR_044037.1 Bacillus coahuilensis m4-4 strain m4-4 2165 94% 0,0 93%
T
NR_025446.1 Bacillus halodurans strain DSM 497 2244 97% 0,0 93%
T
NR_040849.1 Bacillus wakoensis strain N-1 2242 97% 0,0 93%
T
NR_042338.1 Bacillus aerius strain24K 2215 96% 0,0 93%
T
NR_044193.1 Bacillus ginsengi strain ge14 2230 97% 0,0 93%
T
NR_042974.1 Bacillus cibi strain JG-30 2206 97% 0,0 93%
T
NR_040848.1 Bacillus hemicellulosilyticus strain C-11 2213 97% 0,0 93%
 T
NR_042905.1 Bacillus fortis strain R-6514 2268 99% 0,0 93%
T
NR_025473.1 Bacillus decolorationis strain LMG 19507 2202 97% 0,0 93%
T
NR_029022.1 Bacillus indicus strain Sd/3 2204 97% 0,0 93%
T
NR_042084.1 Bacillus murimartini strain LMG 21005 2200 97% 0,0 93%
T
NR_041523.1 Bacillus coagulans strain NBRC 12583 2159 95% 0,0 93%
T
NR_044192.1 Bacillus beijingensis strain ge10 Ba 2207 97% 0,0 93%
T
NR_026145.1 Bacillus pseudalcaliphilus DSM 8725 2202 98% 0,0 92%
T
NR_042648.1 Bacillus cecembensis strain PN5(T) 2187 97% 0,0 92%
T
NR_043013.1 Bacillus alveayuensis strain TM1 2209 98% 0,0 92%
T
NR_036940.1 Bacillus lehensis strain MLB2 2194 98% 0,0 92%
T
NR_026143.1 Bacillus gibsonii strain DSM 8722 2176 97% 0,0 92%
T
NR_026140.1 Bacillus clausii strain DSM 8716 2169 97% 0,0 92%
T
NR_036889.1 Bacillus alcalophilus strain 1 2167 97% 0,0 92%
T
NR_024798.1 Bacillus krulwichiae strain AM31D 2185 97% 0,0 92%
T
NR_027227.1 Bacillus infernus strain TH-23 2161 97% 0,0 92%
T
NR_042217.1 Bacillus arsenicus strain Con a/3 2159 97% 0,0 92%
T
NR_025258.1 Bacillus odysseyi strain 34hs1 2146 97% 0,0 92%
T
NR_043402.1 Virgibacillus pantothenticus strain IAM 11061 2082 95% 0.0 92%
T
NR_025595.1 Bacillus gelatini strain LMG 21880 2130 95% 0.0 92%
T
NR_043021.1 Geobacillus thermodenitrificans strain BGSC 94A2 2182 100% 0.0 92%
T
NR_040794.1 Geobacillus stearothermophilus strain IFO12550 2087 97% 0.0 91%
NR_029304.1 T
Halobacillus litoralis strain SL-4 2152 100% 0.0 91%
NR_037100.1 Anoxybacillus pushchinoensis strain k-1 1842 86% 0.0 91%
Supplementary data Table S2. Values obtained for conserved DNA, ANI and Index similarity SIM for
all pairwise whole-genome sequence comparisons, calculated as previously described (Goris et al.,
2007).

Data are mean of reciprocal values. A, conserved DNA (%) ; B, ANI (%); C, Index SIM calculated as:
(ANI value) x (conserved DNA value)/100.

Strain 1 2 3 4 5 6 7 8

A
1. B. cytotoxicus sp. Nov. NVH 391-98
T
100,0
2. B. pseudomycoides DSM12442
T
23,8 100,0
3. B. weihenstephanensis KBAB4 18,4 20,0 100,0
4. B. mycoides DSM 2048
T
17,7 20,0 75,3 100,0
5. B. cereus ATCC 14579
T
17,3 18,3 59,5 63,1 100,0
6. B. thuringiensis 10792
T
17,4 17,6 59,7 63,1 85,2 100,0
7. B. anthracis A0488
T
16,3 16,0 57,0 58,4 70,9 62,2 100,0
8. B. subtilis NCBI 3610
T
1,0 0,1 0,8 0,1 0,7 0,0 0,3 100,0

B
1. B. cytotoxicus sp. Nov. NVH 391-98
T
100,0
2. B. pseudomycoides DSM12442
T
88,6 100,0
3. B. weihenstephanensis KBAB4 88,7 89,0 100,0
4. B. mycoides DSM 2048
T
88,6 89,1 97,9 100,0
5. B. cereus ATCC 14579
T
88,9 88,9 90,8 90,7 100,0
6. B. thuringiensis 10792
T
88,7 88,9 90,8 90,7 97,0 100,0
7. B. anthracis A0488
T
88,8 88,7 90,8 90,6 92,4 92,1 100,0
8. B. subtilis NCBI 3610
T
- - - - - - - 100,0

C
1. B. cytotoxicus sp. Nov. NVH 391-98
T
100,0
2. B. pseudomycoides DSM12442
T
21,1 100,0
3. B. weihenstephanensis KBAB4 16,3 17,8 100,0
4. B. mycoides DSM 2048
T
15,6 17,8 73,7 100,0
5. B. cereus ATCC 14579
T
15,4 16,2 54,0 57,3 100,0
6. B. thuringiensis 10792
T
15,4 15,7 54,2 57,3 82,6 100,0
7. B. anthracis A0488
T
14,5 14,2 51,8 52,9 65,5 57,2 100,0
8. B. subtilis NCBI 3610
T
<1 <1 <1 <1 <1 <1 <1 100,0
Supplementary data Table S3. Fatty acid composition of B. cytotoxicus strains, B. cereus Group type strains and B. subtilis.

Data are % of total fatty acids and gave similar results during two experiments (mean of standard deviations from 0.03 to 1.46%).

antei
antei antei iso- iso- antei iso- iso- C18:2
iso- iso- iso- iso- iso- C16:1 C16:1 iso- so- C18:1
Strain C12:0 so- C14:0 so- C15:0 C16:1 C16:1 C16:0 so- C17:1 C17:1 C17:0 C18:0 ω-6
C12:0 C13:0 C14:0 C15:0 C16:0 ω-11 ω-6 C17:0 C17:0 C17:1 ω-9
C13:0 C15:0 ω-10 ω-5 ω-11 ω-6
ω-6
T
B. cytotoxicus NVH 391-98 0,4 0,6 7,0 1,8 5,0 2,4 36,5 10,8 0,1 6,7 0,1 0,8 10,8 0,0 3,3 8,2 3,4 0,1 0,7 0,1 0,5 0,5 0,1 0,0
B. cytotoxicus INRA AF2 0,4 0,7 6,9 1,9 4,3 2,7 37,9 9,9 0,1 6,0 0,1 0,8 10,6 0,0 3,8 8,0 3,4 0,1 1,0 0,1 0,6 0,4 0,2 0,1
B. cytotoxicus NVH883/00 0,4 0,6 5,6 1,7 4,3 3,8 40,2 11,3 0,2 6,2 0,2 1,7 7,9 0,0 5,5 5,0 3,0 0,1 1,2 0,0 0,8 0,2 0,1 0,1
B. cytotoxicus CVUAS2833 0,3 0,8 6,1 1,4 3,4 3,6 39,6 9,5 0,2 5,2 0,1 0,7 11,7 0,0 4,3 7,5 3,2 0,1 0,9 0,1 0,7 0,5 0,1 0,1
B. cytotoxicus 08CEB44BAC 0,4 0,7 6,4 1,5 4,7 4,5 39,4 8,6 0,2 6,2 0,0 0,9 11,8 0,1 4,2 6,0 2,5 0,0 0,8 0,1 0,4 0,4 0,1 0,1
T
B. cereus ATCC 14579 1,9 1,4 20,3 4,0 4,8 4,1 20,2 6,5 0,3 3,3 0,2 0,7 12,5 0,5 5,9 6,7 1,5 2,7 1,0 0,2 0,6 0,5 0,1 0,1
T
B. thuringiensis CIP 53137 1,5 1,0 18,5 2,8 5,2 4,1 21,8 5,3 0,8 3,7 0,2 1,3 10,3 0,3 7,5 6,9 1,1 3,2 2,6 0,3 0,5 0,6 0,2 0,0
T
B. weihenstephanensis WSBC 10204 2,9 2,5 22,3 5,8 3,5 3,6 12,6 5,4 0,1 2,9 0,4 0,2 18,0 1,1 3,6 6,6 1,7 4,2 0,4 0,2 0,4 1,3 0,1 0,1
B. weihenstephanensis KBAB4 3,8 2,8 21,7 7,4 4,1 3,5 10,3 5,9 0,2 3,1 0,4 0,5 14,2 1,0 4,8 6,4 1,9 5,1 0,8 0,2 0,6 1,4 0,1 0,1
T
B. mycoides DSM 2048 2,6 2,7 21,9 3,9 3,4 3,7 12,5 3,8 0,2 2,8 0,4 0,5 15,6 1,2 6,4 7,5 1,1 6,5 1,0 0,1 0,4 1,6 0,1 0,0
T
B. pseudomycoides DSM12442 8,7 1,4 12,6 4,9 5,5 3,2 13,3 3,6 0,9 8,3 0,1 2,9 9,0 0,0 12,3 7,0 1,6 0,1 2,3 0,6 1,1 0,5 0,1 0,1
T
B. subtilis LMG 7135 0,1 0,8 0,2 0,1 1,9 1,1 9,1 40,3 0,4 6,4 0,7 0,3 8,1 1,5 0,1 7,5 13,5 3,5 0,6 0,4 0,7 2,3 0,1 0,2
B. subtilis 168 0,0 0,1 0,2 0,1 1,6 0,5 18,5 37,5 0,1 4,9 0,2 0,0 8,2 0,6 0,0 13,1 11,7 1,7 0,1 0,1 0,0 0,8 0,0 0,0
Supplementary Figure S1. Relationship between experimental DDH values and genomic
identity (ANI), conservation (conserved DNA and Index of similarity SIM) between
genomes. ANI and conserved DNA presented here were calculated as previously described
(Goris et al., 2007) at a threshold of 85% identity. The index SIM was derived from the two
previous parameters [(ANI% x conserved DNA%) threshold 85% /100] and thus take into account
the two parameters. Among the thresholds of 70%, 75%, 80%, 85% and 90%, the
thresholds of 85% was shown to generate the best compromise for correlation level
between our experimental DDH values and the % of conserved DNA and for a linear
regression model from the ideal situation (y=x). Here, at a thresholds of 85% identity the
Index SIM generate the best correlation with our experimental DDH values through a linear
regression model (r2=0.947, y close to x). This was thus the chosen model to predict DDH
values from all pairwise (whole-genome) sequence: y=1.066 x – 1.9689. Note that the
chosen thresholds of 85% identity and the chosen model depends on the experimental
DDH data that may be obtained from different DDH conditions, in particular from different
stringency conditions. Our choice here reflects our experimental data/conditions.

●, DDH value between two strains plotted against ANI value calculated from the two same
strains; …….. confidence interval (Mean 95%); ____ confidence interval (Obs. 95%), ___ Model

R²=0,933
y = 6,8731 x - 577,5 (R²=0,740) y = 1,0277 x - 4,2347
Pearson=0,966
140
Pearson=0,860 100

120
80
100
60
80
DDH

DDH

60 40

40 20
20
0
0 10 30 50 70 90
88 90 92 94 96 98 -20
Conserved DNA
ANI

R²=0,947
y = 1,066x - 1,9689
coeff Pearson=0,973
120

100

80

60
DDH

40

20

0
10 30 50 70 90
-20
Index SIM