FEMS Microbiology
Letters 128 (1995) 113-118
Differentiation of Bacillus anthracis and other ‘Bacillus
group’ bacteria using IS231 -derived sequences
Ian Henderson a,*, Yu Dongzheng
cereus
b, Peter C.B. Turnbull a
a Centre for Applied Microbiology and Research, Porton Down, Salisbury, Wiltshire SP4 OJG, UK
h Institute of Epidemiology and Microbiology, P.O. Box 5, Changping, Beijing 102206, People’s Republic of China
Received
10 February
1995; revised 2 March 1995; accepted 7 March 1995
Abstract
Sequences based on the conserved 20 bp inverted repeat of IS231 variants were used as polymerase chain reaction-based
fingerprinting primers of the member species of the Bacillus cereus group (B. anthracis, B. cereus, B. thuringiensis and B.
mycoides), because of their close association with transposons, principally Tn4430 in B. thuringiensis. Fingerprints of B.
anthracis were simple, and specifically allowed its identification and sub-differentiation
from other members of the group.
Fingerprints for B. cereus were strain-specific; those for B. thuringensis gave a 1650 bp product, characteristic of IS231
variants A-F. The same reaction conditions gave one or two bands for both f?. anthrucis and B. cereus that differed by
restriction endonuclease mapping from the B. thuringiensis PCR product and established IS231 restriction maps; this does
not preclude some kind of relationship between these products and IS231.
Keywords:
Bacillus
Bacillus cereus; Bacillus thuringiensis; IS231; Polymerase
anthracis;
1. Introduction
The
cereus,
‘Bacillus
cereus
B. thuringiensis
group’,
B. anthracis,
B.
B. mycoides,
are notable for their phenotypic
relatedness. The species
and subspecies of the group can be differentiated
only on the basis of highly mutable characteristics
such as colony morphology, penicillin and gamma
phage susceptibility, motility and lack of haemolysis,
and the elaboration of certain virulence factors. Avirulent environmental
isolates of B. anthracis are only
distinguishable
from B. cereus if they retain penicillin and gamma sensitivity, are weakly haemolytic
* Corresponding
610898.
author.
and
Tel:
01980
0378-1097/95/$09.50
0 1995 Federation
SSDI 0378.1097(95)00092-5
612361;
Fax:
of European
019801
Microbiological
chain reaction differentiation
and are non-motile.
This indicates the need for a
suitable means of differentiating
between members
of the group. The increasingly frequent identification
of avirulent B. anthracis in the environment [I], and
the strategic interest in the organism, underscores a
need not only for a system of rapid identification, but
also for a strain differentiation
and epidemiological
tracing capability.
Probes of chromosomal organisation seem to be
the most likely route to fulfill these requirements.
We have recently identified the randomly amplified
polymorphic DNA polymerase chain reaction (RAPD
PCR) technique to be capable of at least specifically
identifying B. anthracis from other B. cereus group
members [2]. Slight differences
between
B. anthracis
isolates
Societies.
All rights reserved
were
detected
using
this
method
I. Henderson el al. / FEMS Microbiology Letters I28 (I 995) 113-l 18
114
when pure chromosomal
DNA was used, but the
reproducibility
of the technique
is likely to be
severely challenged when complex environmental
samples are presented such as soil, due mostly to the
sensitivity of PCR to cation concentrations
and the
presence of phenolics. The alternative is to simplify
PCR fingerprint patterns, and to aim for targets that
are more likely to have some influence on chromosoma1 organisation,
such as insertion sequences and
transposons.
One of the best characterised of these types of
sequence in the B. cereus group is IS231. At least
eight variants (A-F, V and W) of this sequence exist
in B. thuringiensis [3-51, and have been shown to be
related to the IS4 family in Escherichia coli [6].
They are delimited by an 11 bp street repeat sequence and a highly conserved 20 bp inverted repeat
sequence, and have been shown to be associated with
Tn4430 in B. thuringiensis strain berliner 1715 [7].
Such sequences have yet to be characterised in B.
anthracis and B. cereus although Tn4430 has been
shown to facilitate transfer of the B. anthrucis virulence encoding plasmids pXO1 and pXO2 [8,9] between members of the B. cereus group.
The purpose of this study was to use PCR primers
based on the IS232 20 bp direct repeat as determinants of species-to-species
and isolate-to-isolate
variation in the members of the B. cereus group, and
specifically in B. anthrucis. It was not the purpose to
isolate further IS231 variants from any of the organisms examined but clearly some of the PCR products
produced may represent such elements.
2. Materials
2.1. Bacterial
and methods
Table I
Sources of ‘Bacillus cereus group’ isolates
Organism
a.’
Other designation
Host/Source
B. anthracis:
NCTC 8234
Sterne
ASC 182
ASC 328
Pasteur
Vellum
ASC 184
ASC 185
ASC 68
Vellum e
Vollum e
Ames
ASC 69
New Hampshire
cow, 1937
South Africa ’
France, pre-1880’s
Cow, pre-1939
United Kingdom
pxo1+ pxo2pxo1pxo2+
Cow, 1980
USA
Human, 1957
USA
Human, 1982
Zimbabwe
pxo1pxo2+,
penicillin-resistant
isolate ’
Elephant, 1983
Namibia
As ASC 58
ASC 52
ASC 183
ASC 58
ASC 60
’
Reference
11
12
13
14
15
16
a Other B. cereus group strains used in this study: B. thuringiensis
HD37, HD102 and HD225, obtained from H.T. Dulmage, US
Department
of Agriculture,
Brownsville,
Texas,
USA; B.
thuringiensis F2113/78
(ssp. entomocidus); B. cereus NCTC
2599, F4810/72,
F2532/74,
F3484/77,
F4433/73, ASC 109,
ASC 112 and ASC 113; E. mycoides NCTC 09680. All strains
with an F prefix were obtained from the Food Hygiene Laboratory. Central Public Health Laboratory, Cohndale, London, UK.
h ASC: Anthrax Section Culture, Research Division, CAMR, Porton Down, Salisbury, UK.
’ Unless
otherwise
stated,
B. anthracis
isolates
are
pxo1+ pxo2+.
d Livestock and UK human vaccine strain.
’ Kindly supplied by Defence Microbiology
Division, Chemical
and Biological Defence Establishment,
Porton Down, Salisbury,
UK.
f ASC 32 cured by Dr. C.P. Quinn, Molecular Microbiology
Group, CAMR, Porton Down, Salisbury, UK.
strains and isolates
.B. anthracis, B. cereus, B. thuringiensis and B.
mycoides isolates and strains are presented in Table
1. Chromosomal
DNA was isolated by a method
described elsewhere 121.
2.2. PCR
PCR primers (Table 2) were synthesized by the
automated phosphoramidite
method using an Applied Biosystems model 380B DNA synthesizer. PCR
mixtures of 100 ~1 consisted of 1.25 mM each of
dATP, dCTP, dGTP and dTTP, 3.0 mM MgCl,, 2
PM of single PCR primer, and 10 ~1 of 10 X
reaction buffer (160 mM (NH,),SO,,
670 mM TrisHCl (pH 8.81, 0.1% Tween-20). PCR cycling conditions consisted of 95°C for 5 min and 30 cycles of
94°C for 2 min, 40°C for 1 l/2 min, 72°C for 2 min
(primer 1, Table 2) or 3 min (primer 2, Table 21, and
finally 72°C for 5 min, in an MJ Research thermocycler. PCR products were analyzed by agarose gel
electrophoresis [ 101.
I. Henderson et al. / FEMS Microbiology Letters 128 (1995) 113-I 18
from B. thuringiensis
var israeliensis. The upper
band had a molecular mass in excess of 3.1 kb and
was produced only from certain isolates. The presence of this extra band is independent of virulence.
B. cereus has a much greater pattern variety compared to both B. thuringiensis
and B. anthracis.
Among these varieties are two B. cereus strains
which appear identical to B. thuringiensis (B. cereus
F4810/72
Fig. lA, lane 7, and F564/49
Fig. lA,
lane 6), and one identical to the B. anthracis double
band pattern (B. cereus F3484/77 Fig. lB, lane 9).
These findings suggest the possibility
that all
three species and sub-species may harbour variants
of IS231-like sequences. To examine this, limited
restriction mapping analysis was performed on the
1.6 kb band of B. cereus and B. thuringiensis and
the 1.9 kb band of B. anthracis. These comparisons
are presented in Fig. 2. Although direct comparisons
between known and newly determined
restriction
maps are difficult without sequence data, one can see
from these maps that known IS231 variants have
highly similar site distributions (and hence conserved
DNA sequences). This is borne out by alignments of
the DNA sequences of IS231 variants [3-51. There
appears to bo good correlation
B: thuringiensis
HD102 PCR product and B. cereus ASC 109 with
established restriction maps of IS231 variants. There
was no correlation between the B. anthracis 1.9 kb
band map and these variants (Fig. 2).
Of particular interest to this study are the fingerprints generated with the PCR primers. Fingerprints
with primer 1 were not species-specific;
certain patterns were common to B. thuringiensis and B. cereus
2.3. PCR product characterisation
To analyze the PCR products by restriction endonuclease mapping, samples were separated in low
melting point agarose (Sigma) and cut from the gel;
the DNA was subsequently
purified using Wizard
minicolumns
(Promega). A restriction map of each
band was determined using several restriction endonucleases. These enzymes were used according to
manufacturer’s
instructions.
Results and discussion
Primers 1 and 2 (Table 2) provide the means to
assess (i) the possibility
of IS231-like
sequences
being present in the genomic DNA of B. anthracis
and (ii) variability
in the spacing between these
IS23I-like
sequences in the genome, respectively.
Using the 20 bp inverted repeat sequence as a PCR
target means that only a single primer needs to be
used in each PCR reaction allowing easy optimisation of the reaction.
Fig. 1 shows PCR profiles for primer 1. Reactions
were optimised by varying primer annealing temperature and the magnesium ion concentration.
All the
B. thuringiensis strains analyzed produced a single
band of approximately
1650 bp, the characteristic
size of IS231 A-F variants (Fig. 1A). PCR products
from B. anthracis gave characteristic
one or two
band patterns (lanes 2-13, Fig. 1B). The lower band
has a,molecular mass of approximately 1.9 kb, within
the correct size range of IS231 variants V and W
Table 2
Terminal
inverted repeats of IS231 variants
IS.231
Variant
Sequence
A
B
C
D
E”
F
V
W
Primer 1
Primer 2
S-CAT
5’.CAT
S-CAT
5’.CAT
5’-CAT
5’-CAT
5’-CAT
S-CAT
5’-CAT
3’-GTA
’ This sequence
and primers used for PCR
Source/Reference
GCC
GCC
GCC
GCC
ACC
GCC
CGC
CGC
GCC
CGG
CAT
CAT
CAT
CAT
CAT
CAT
CAT
CAT
CAT
GTA
CAA
CAA
CAA
CAA
CAA
CAA
CAA
CAA
CAA
G’IT
is only found at the 3’ end of 1.9231.
Cl-T
CTT
CIT
ATT
CIT
Cl-f
GCT
GCT
CTI
FAA
AAG
AAG
AAG
AAA
AAG
AAG
AAG
AAG
AAG
TK
11s
AA-3’
AA-3’
AA-3’
GA-3’
GG-3’
AA-3’
GA-3’
GA-3’
AA-3’
‘IT-5’
5
This study
This study
116
I. Henderson et al. /FEMS Microbiology Letters 128 (1995) 113-118
4072bp
2036bp
1635bp
1018bp
0
123
=bp
1635bp
1018bp
Fig. 1. PCR fingerprints generated using primer 1. Isolates and
strains are (A) 1 kb ladder (l), E. fhuringiensis HD37 (2), HD102
(3), HD225 (4), F2113/78
(S), 8. cereus 564/49 (6) F4810/72
(7) F2532/74 (8), F3484/77
(9), NCTC 2599 (IO), B. mycoides
NCTC 09680 (11) and 1 kb ladder (12). (B) B. anthracis NC’K
8234 (1) AX 11 (2), ASC 58 (3), ASC 68 (4), ASC 69 (5), ASC
328 (6) ASC 184 (7) ASC 185 (8), ASC 81 (9), ASC 182 (lo),
ASC 183 (ll), ASC 55 (12), ASC 60 (13) 8. cereus ASC 109
(14) ASC 112 (15), ASC 113 (16) 1 kb ladder (GibcoBRL) (17).
Electrophoresis was carried out in 0.8% (w/v) agarose for 16 h at
in the laboratory, where it appears to be reluctant to
grow on B. anthracis selective media and in horse
blood (for capsule production tests). This difference
is seen further with primer 2 in the next section, and
confirmed
using rRNA fingerprinting
approaches
(Henderson,
I. and Duggleby, C.J., manuscript in
preparation).
Primer 2 fingerprint patterns were different for
each strain of B. cereus, B. thuringiensis and B.
mycoides tested and therefore could not be used for
their differentiation.
In contrast, patterns for B. anthracis were highly specific. Only three pattern variants could be detected for all of the B. anthracis
isolates tested (Fig. 3); these did not overlap with
patterns
for B. cereus,
B. mycoides
and B.
thuringiensis. Variation was found principally in the
presence or absence of the PCR product of approximately 1.6 kb. Those isolates that do not have this
band also lack the 3.1 kb band with primer 1. ASC
182 also lacks this band and has a different fingerprint with primer 2 compared to other B. anthracis
isolates, confirming the suspected difference of this
isolate [3]. This difference may be as a result of
numerous laboratory manipulations
in the > 100
years since it was first isolated. As with primer 1, the
reasons for the differences observed with primer 2
are unclear, showing no relationship
with documented histories. We must conclude, therefore, that
possibly with the exception of ASC 182, these differences at the level of chromosomal DNA appear to be
at random in the absence of other data.
Consistency of DNA fingerprints derived by any
technique does not mean that isolates are identical.
Although a high degree of band-sharing does occur
55 v.
000
lsssbp
D
165=Q
lA), and B. cereus and B. anthrucis (Fig. 1B).
This primer could not be used to differentiate between this group of closely related species. The
differences observed between B. anthracis isolates
do not correspond to the known documented histories of these isolates. For example ASC 184 (Fig.
lB, lane 7) and ASC 185 (Fig. lB, lane 8) are from
a common ancestor yet give different fingerprint
patterns. However, ASC 182, which only gives a
single band (Fig. lB, lane lo), is seen to be different
(Fig.
lassbp
lssmp
lassbp
lassbp
-SW
lssobp
lewbp
lombp
D
D
0
0
0
I
D
0
D
OD
D
D
00
0
I
0
D
D
IS231A
0
DD
D
0
IS2376
IS231C
IS237D
IS237E
IS237F
IS23N
HD102
D
I
0
D
D
1
0
B.cweus
8. anthrads
Fig. 2. Alignment of Dral (D) restriction enzyme sites for known
IS231 variants and PCR products of B. thuringiensis HD102, B.
cereus ASC 109 and all B. anthracis.
I. Henderson et al. / FEMS Microbiology Letters 128 (1995) 113-I 18
with the products from the two specific primers in
this study for B. anthmcis, the individual bands
themselves may differ at the DNA sequence level.
To this end, bands of equal electrophoretic mobility
generated from both primer 1 and 2 were isolated
and purified from low melting point agarose for
several of the B. anthracis isolates analysed. Cloning
and sequencing was beyond the scope of this study,
but isolated DNA’s were restriction-mapped.
Typical
results are presented in Fig. 4. Bands of comparable
electrophoretic
mobility were identical at this level
of analysis. Isolate-to-isolate
variation was not observed, reinforcing
the notion of the highly conserved nature of B. anthracis.
In summary, sequence-specific
or sequence-directed oligonucleotide
primers are useful fingerprinting tools for the B. cereus group. This is particularly
the case for B. anthracis where the high degree of
conservancy between isolates and strains makes fingerprinting the best approach for identification
and
sub-differentiation.
A similar conclusion was drawn
from Ml3 repeat sequence fingerprinting
using PCR
[2] but those fingerprints were much more complex
and difficult to reproduce. Fingerprints
in this instance are simple and reproducible, with differences
1 2 3 4 5 6 7
1 2 3 4 5 6 7
117
8 910111213141516
4072bp
2036bp
1635bp
1018bp
500bp
Fig. 4. Restriction mapping of the 1.9 kb PCR product of B.
anthracis ASC 91 and ASC 182 generated with primer 1. Lanes
are: 1 kb ladder (l), ASC 91 DraI (21, DraI/XbaI (3). XbaI (4),
XbaI/SspI
(51, SspI (61, DraI/SspI
(71, ASC 182 DraI (81,
DraI/XbaI (9), XbaI(lO), XbaI/SspI
(111, SspI (121, DraI/SspI
(131, and 1 kb ladder (14). Enzymes that did not cut the PCR
products were BamHI, BscI, CfoI, EcoRI, EcoRV,
HaeII,
HindIII, Mlul, NraI, PstI, Scul, StuI and XhoI. Digested DNA
was analysed by electrophoresis
in 0.8% (w/v) agarose for I6 h
at 55 V.
8 910111213141516
3054bp
2036bp
1635bp
1016bp
5OObp
Fig 3. PCR fingerprints generated using primer 2. 1 kb ladder (11,
B. anthracis NCTC 8234 (2), ASC 11 (31, ASC 58 (4), ASC 68
(51, ASC 69 (6), ASC 328 (71, ASC 184 (81, ASC 185 (91, ASC
327 (101, ASC 91 (111, ASC 182 (121, ASC 183 (131, ASC 55
(14), ASC 60 (151, and 1 kb ladder (16). Electrophoresis
was
carried out in 0.8% (w/v) agarose for 16 h at 55 V.
between individual B. anthracis isolates being much
more obvious.
Fingerprint variation between B. anthracis isolates is limited to gross band differences. Differences
between bands of comparable electrophoretic mobility do not seem to be apparent although extensive
DNA sequencing of such PCR products needs to be
performed to confirm this. The life cycle of B.
anthracis depends on its finding a suitable host in
which to multiply; in the environment
it exists almost invariably in the spore form. The opportunity
for influences of DNA variation, such as exposure to
phages and other DNA transforming events, is therefore limited to the relatively infrequent times when
the organism is in the vegetative stage. Fingerprinting methods using specific targets show promise for
studies of the variation of the chromosome for B.
anthracis as opposed to those which probe the chromosome in a random manner.
118
I. Henderson et al. /FEMS
Microbiology Letters I28 (1995) 113-I 18
Acknowledgements
The authors thank the Royal Society for the Study
Visit award to Dr. Yu which made his contribution
to this research possible. The technical assistance of
Ms. Caroline Redmond is also gratefully acknowledged.
References
[l] Turnbull, P.C.B., Hutson, R.A., Ward, M.J., Jones, M.N.,
Quinn, C.P., Finnie, N.J., Duggleby, C.J., Kramer, J.M. and
Melling, J. (1992) Bacillus anrhracis but not always anthrax.
J. Appl. Bacterial. 72, 21-28.
[2] Henderson, I., Duggleby, C.J. and Turnbull, P.C.B. (1994)
Differentiation
of B. anthracis from other Bacillus cereus
group bacteria with the PCR. Int. J. Sys. Bacterial. 44,
99-105.
[31 Mahillon, J., Seurinck, J., Delcour, J. and Zabeau, M. (1987)
Cloning and nucleotide sequence of different iso-IS231 elements and their structural association with Tn4430 transposon in Bacillus thuringiensis. Gene 51, 187-196.
141 Rezsohazy, R., Hallet, B. and Delcour, J. (1992) IS231D, E
and F, three new insertion sequences in Bacillus thuringiensis: extension of the IS231 family. Mol. Microbial. 6, 19591967.
[51 Rezsohazy, R., Hallet, B., Mahillon, J. and Delcour, J.
(1993) IS231V and W from Bacillus thuringiensis, two
distant members of the IS231 family of insertion sequences.
Plasmid 30, 141-149.
id Mahillon, J., Seurinck, J., Van Rompuy, L., Delcour, J. and
Zabeau, M. (1985) Nucleotide sequence and structural organ-
isation of an insertion sequence element (IS231) from Bacil/us thuringiensis strain berliner 1715. EMBO J. 4, 38953899.
171 Mahillon, J. and Lcreclus, D. (1988) Structural and functional analysis of Tn4430: identification of an integrase-like
protein involved
in the co-integrate-resolution
process.
EMBO J. 7, 1515-526.
k31Ruhfel, R.E., Robillard, N.J. and Thorne, C.B. (1984) Interspecies transduction of plasmids among Bacillus anthracis,
B. cereus, and B. fhuringiensis. J. Bacterial. 157, 708-711.
[91 Green, B.D., Battisti, L. and Thorne, C.B. (1989) Involvement of Tn4430 in transfer of Bacillus anthrucis plasmids
mediated by Bacillus thuringiensis plasmid pXO12. J. Bacteriol. 171, 104-113.
1101 Sambrook, J., Fritsch, E.F. and Maniatis, T. (1989) Molecular Cloning: a Laboratory
Manual, 2nd ed. Cold Spring
Harhour Laboratory Press, Cold Spring Harbour, NY.
[ill Sterne, M. (1937) The effects of different carbon dioxide
concentrations
on the growth of virulent anthrax strains.
Pathogenicity and immunity tests on guinea-pigs and sheep
with anthrax variants derived from virulent strains. Onderstepoost J. Vet. Sci. Anim. Ind. 9, 49-67.
[121 Pasteur, L. (1881) De l’attenuation des virus et de Ieur retour
a la virulence. C.R. Acad. Sci. 92, 429-435.
[131 Smith, N.E., Gordon, R.E. and Sneath, P.H.A. (1964) Type
cultures and proposed neotype cultures of some species in
the genus Bacillus. J. Gen. Microbial. 34, 269-272.
[I41 Plotkin, S.A., Brachman, P.S., Utell, M., Bumford, F.H. and
Atchisun, M.M. (1960) An epidemic of inhalation anthrax,
the first in the twentieth century. Am. J. Med. 29, 992-1001.
1151 Davies, J.C.A. (1982) A major epidemic of anthrax in Zimbabwe. 1. Cent. Afr. J. Med. 28, 291-298.
h1 Turnbull, P.C.B., Hofmeyr, J.M., McGetrick, A.M.T. and
Oppenheim, B.A. (1986) Isolation of Bacillus anthracis, the
agent of anthrax, in the Etosha National Park. Madoqua. 14,
321-331.