Professional Documents
Culture Documents
Sebei Et Al-2007-Journal of Applied Microbiology
Sebei Et Al-2007-Journal of Applied Microbiology
Sebei Et Al-2007-Journal of Applied Microbiology
ORIGINAL ARTICLE
Keywords
antimicrobial peptide, Bacillus cereus,
bacteriocin, purification, sequencing.
Correspondence
S. Sebei, Laboratory of Microbial Technology,
Division of Microbial Science and Technology,
Department of Bioscience and Biotechnology,
Faculty of Agriculture, Graduate School,
Kyushu University, 6-10-1 Hakozaki,
Higashi-ku, Fukuoka 812-8581, Japan.
E-mail: sebeisamir@yahoo.fr
Abstract
Aim: To purify and characterize the bacteriocin produced by strain MRX1.
Methods and Results: A bacteriocin-producing strain was isolated and identified as Bacillus cereus. The bacteriocin, called cerein MRX1, was purified from
the culture supernatant using hydrophobic interaction, cation-exchange chromatography and RP-HPLC. It could also be purified in abundance from the
cell surfaces of the producer strain. Mass spectrometry revealed its molecular
mass of 313793 Da. Sequencing of chemically modified bacteriocin identified
its partial sequence: DWTCWSCLVCAACSVELL. Amino acid analysis, confirmed by 1H-NMR, suggested cerein MRX1 to be a class II bacteriocin. This
bacteriocin was remarkably hydrophobic, heat-stable and could withstand a
wide range of pH. It exhibited a bactericidal mode of action against Bacillus
coagulans JCM 2257T. Cerein MRX1 was especially active against spoilage bacteria such as Bacillus subtilis and Listeria innocua (MICs in the 1 lg ml)1
range). In contrast, lactic acid bacteria were resistant or required higher concentrations to be inhibited.
Conclusions: Cerein MRX1 is similar by its N-terminal sequence to thuricin 17
recently isolated from Bacillus thuringiensis NEB17. However, the two bacteriocins are different by their molecular masses and amino acid compositions.
Significance and Impact of the Study: Chemical stability of cerein MRX1 and
its ability to inhibit a large number of undesirable bacteria may give an advantage to its food or clinical application as an antibacterial agent.
Introduction
Bacteria produce a variety of antimicrobial substances that
are able to kill or inhibit other micro-organisms. Bacterial
antibiotics can be subdivided into two types on the basis
of their chemical nature: (i) nonpeptide antibiotics such
as the antibacterial and antifungal phospholipid bacilysocin produced by Bacillus subtilis 168 (Tamehiro et al.
2002); and (ii) peptide antibiotics, which fall into two
large groups differentiated on the basis of the biosynthetic
pathways by which they are generated. One group comprises nonribosomally synthesized peptides produced on
large enzymatic complexes. These peptides consist of only
1621
S. Sebei et al.
Bacterial identification
Sugar fermentation profiling of strain MRX1 was carried
out by means of API 50 CHB (bioMerieux, Marcy
lEtoile, France). Total DNA was extracted according to
Kalman et al. (1993). Two primers were used for 16S
rDNA analysis: MRXF1: 5-GCGGCGTGCCTAATACATGC-3 and MRXR2: 5-CTCTACGCATTTCACCGCTAC-3. PCR was performed using Ex Taq polymerase
(Takara, Otsu, Japan) under the following conditions:
94C for 3 min followed by 30 cycles of denaturation at
94C for 30 s, annealing at 56C for 30 s and polymerization at 72C for 1 min. Two primers (Manzano et al.
2003) were used for gyrB gene analysis: BCFW1: 5GTTTCTGGTGGTTTACATGG-3 and BCRW1: 5-CAACGTATGATTTAATTCCACC-3. PCR was performed
using KOD plus polymerase (Toyobo, Osaka, Japan)
under the following conditions: 94C for 2 min followed
by 30 cycles of denaturation at 94C for 30 s, annealing
at 50C for 30 s and polymerization at 68C for 30 s. All
PCR were performed in a T-GRADIENT thermocycler
(Biometra, Gottingen, Germany). The PCR products were
purified by Qiagen PCR purification kit (West Sussex,
UK), ligated into pGEM-T vector (Promega, Madison,
WI, USA) and subjected to DNA sequencing.
Bacteriocin purification
Cerein MRX1 was produced in a 1-l culture of TSBYE
(tryptic soy broth plus 06% yeast extract, Becton Dickinson) inoculated with 10 ml of strain MRX1 precultured for
24 h at 30C with shaking. The 1-l culture was carried out
in a 3-l Erlenmeyer flask for 20 h with shaking.
Subsequently, cells were harvested by centrifugation
(10 000 g, 15 min, 4C) and the antimicrobial activity was
concentrated by hydrophobic interaction chromatography
(HIC) as follows: the supernatant fluid was shaken with
20 g of Amberlite XAD-16 resin (Sigma, St Louis, MO,
USA) for 3 h. The matrix was then packed into a column
(25 by 20 cm, Bio-Rad, Tokyo, Japan) and washed successively with 100 ml of distilled H2O and 100 ml of 40%
ethanol. The bacteriocin was eluted with 200 ml of 70%
2-propanol containing 01% TFA (trifluoroacetic acid).
2-propanol was evaporated at 35C using a vacuum system.
The resulting material was diluted in 100 ml of sodium
phosphate buffer, 20 mmol l)1, pH 57 and subjected to a
20-ml SP-Sepharose Fast Flow cation-exchanger (15 by
14 cm, Amersham Pharmacia Biotech, Uppsala, Sweden)
at a flow rate of 2 ml min)1, using a peristaltic pump
(Micro Tube Pump MP-3, Eyela, Tokyo, Japan). Ten-ml
Fractions were collected during elution of the column with
a stepwise gradient from 0 to 1 mol l)1 NaCl. The active
fraction was diluted with 01% TFA and loaded into a
S. Sebei et al.
Indicator strain*
Medium
Temperature (C)
Inhibition
TSBYE
TSBYE
TSBYE
TSBYE
TSBYE
TSB
TSB
TSB
TSB
TSB
TSBYE
TSB
TSB
TSB
TSB
TSBYE
TSBYE
TSBYE
TSB
TSB
TSB
TSB
TSB
TSB
TSB
TSBYE
TSBYE
TSB
TSB
RCM
30
30
30
30
30
30
30
30
30
30
30
30
30
30
30
37
30
30
30
30
30
30
30
37
30
37
37
30
30
30
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
+
)
)
+
)
+
RCM
BHI
TSBYE
MRS
37
37
30
30
+
+
MRS
MRS
MRS
MRS
MRS
MRS
TSBYE
30
30
37
37
30
30
37
)
)
)
)
)
)
)
TSBYE
TSBYE
TSB
TSB
37
37
37
30
+
+
)
)
The active peak of cerein MRX1 shown in Fig. 1 was used to determine this spectrum.
*ATCC, American Type Culture Collection; CIP, Collection de lInstitut Pasteur, France; CECT,
Spanish Type Culture Collection; CCM, Crechoslovack Collection of Micro-organisms; DSM,
Deutsche Sammlung von Mikro-organismen; IFO, Institute for Fermentation, Osaka, Japan; JCM,
Japanese Collection of Micro-organisms; WBC, Weihenstephan Bacillus collection, Germany.
E. coli JM109 was from Promega. Others strains were laboratory isolates.
TSB, tryptic soy broth; TSBYE, TSB with 06% yeast extract; BHI, brain heart infusion; RCM,
reinforced clostridial medium; MRS, de Man-Rogosa-Sharpe agar. +, Strong inhibition of the
indicator strain; ), no inhibition; , hazy zone of inhibition.
1623
reversed-phase column (Resource RPC, Amersham Pharmacia Biotech) integrated in an LC-10A HPLC system
(Shimadzu, Kyoto, Japan). The column was eluted with a
linear gradient from 0% to 100% acetonitrile containing
01% TFA at a flow rate of 1 ml min)1. The bacterial cells
harvested from the 1-l culture were washed in 100 ml of
10 mmol l)1 sodium phosphate buffer (pH 68) for
15 min, and then resuspended in 20 ml of 01% TFA in
40% acetonitrile. The cell suspension was shaken gently for
15 min, then the cells were discarded by centrifugation
(14 000 g, 15 min, 4C) and the cell surfaces extract (containing the bacteriocin) passed through a syringe filter and
stored at )20C. Fractions obtained during all the purification procedures were assayed for antimicrobial activity by
the spot-on-lawn method (Tagg et al. 1976) using Bacillus
coagulans JCM 2257T as indicator organism. Protein concentration was estimated by the dye binding method
(Smith 1995).
Assay of bacteriocin activity
Bacteriocin activity was assayed by the critical dilution
method. Plates containing solid medium were overlaid
with 5 ml of soft agar lawns containing approx. 5 106
cells of the indicator strain. Portions (10 ll) of serial
twofold dilutions of bacteriocin solutions were spotted
onto these lawns. After incubation at the optimal conditions, the bacterial lawns were examined for growth
inhibitory zones. One unit of antimicrobial activity was
defined as the smallest amount to give a clear zone of
growth inhibition and the reciprocal of the highest dilution which caused such a zone was defined as the number
of antimicrobial units (AU) per millilitre. When the aim
of the experiment was to detect the presence of antimicrobial activity, 10 ll of the solutions was spotted on
the lawns without diluting (spot-on-lawn method).
Direct detection of bacteriocin activity
in polyacrylamide gel
SDSPAGE was curried out on a 16% gel with tricine as
trailing ion (Schagger and von Jagow 1987). The bacteriocin was loaded in duplicate onto the gel. After electrophoresis, the gel was cut vertically. One part of the gel
contained the peptide marker and cerein MRX1, the other
part contained cerein MRX1 only. The former part was
stained to visualize protein bands. For in situ detection of
bacteriocin activity (Bhunia et al. 1987), the second part
of the gel was fixed in a mixture of 2-propanol, acetic acid
and H2O (25 : 10 : 65) for 15 min and rinsed with sterile
H2O for 30 min. The gel was then placed in a Petri dish
and overlaid with 10 ml of soft TSBYE agar containing
100 ll of an overnight culture of B. coagulans JCM 2257T.
1624
S. Sebei et al.
S. Sebei et al.
Other methods
Purification step
Volume
(ml)
Total protein
(mg)
Total activity
(AU)
Specific activity
(102 AU mg)1)
Recovery (%)
Purification
(fold)
Supernatant
HIC
CEX
RP-HPLC
1000
200
10
1
1246
4476
805
0076
300
280
16
12
240
625
1987
1578
100
933
53
40
10
26
82
656
000
000
000
000
1625
S. Sebei et al.
220 nm
1500
1000
500
0
1000
280 nm
500
0
0
10
20
30 min
(a)
1
(b)
2
kDa
169
106
62
34
1626
S. Sebei et al.
MICs
(lg ml)1)
313991
4000
07
09
2
15
35
11
12
08
21
142
285
248
R
R
R
R
Intensity
Indicator strains
2000
0
3000
4000
m/z
Figure 3 MALD-TOF mass spectrometry analysis of cerein MRX1. The
m z value of 313991 corresponding to [M + H]+ indicates a molecular mass of 313891 Da.
1627
S. Sebei et al.
Ala
Asx
Cys
Glx
Gly
Leu
Ser
Thr
Trp
Val
6
1
1
1
1
4
3
2
0
3
(586)
(085)
(076)
(096)
(103)
(400)
(218)
(158)
(000)
(283)
*In parenthesis are the experimental values determined by an automated amino acid analyser after hydrolysis of the peptide with 6 mol l)1
HCl for 24 h at 110C. Under these conditions hydroxy-substituted
residues are partially destroyed and Trp is degraded. The determination of hydroxy-substituted residues takes into account the fact that
the ratios observed for Ser (here 218) and Thr (here 158) are typically
30% and 13% below the real values, respectively (Dunn 1995).
From the Edman sequencing data, it was clear that residues Asx and
Glx correspond to 1 Asp and 1 Glu, respectively.
In addition to the 1 Cys residue determined by the amino acid analysis, Edman sequencing showed the presence of 3 other cysteines.
(ppm)
Figure 4 One-dimensional NMR spectrum of cerein MRX1. The part
of the spectrum that encompasses the vinyl and aromatic regions is
shown. The absence of vinyl protons which give doublet [because of
dehydroalanine (Dha)] or quartet [because of dehydrobutyrine (Dhb)]
resonance peaks in the d = 5269 ppm region of the spectrum
(Jones et al. 1983; Liu and Hansen 1992; Paik et al. 1998) is
an important indicator for the absence of the dehydro amino acids
Dha and Dhb. Thus, cerein MRX1 is not a lantibiotic but a class II
bacteriocin.
S. Sebei et al.
on SP-Sepharose, leading to a loss of activity during sample loading and column washing (before elution by
NaCl). The amino acid composition supports this explanation. Cerein MRX1 contains acidic residues (1 Asp and 1
Glu) but not basic residues (Arg, Lys and His). Therefore,
this bacteriocin is expected to have a net negative charge
(close to )2) over a relatively wide range of pH. It is
important to notice that this feature is rare among bacteriocins from Gram-positive bacteria. Most bacteriocins
are cationic because they contain an excess of basic residues (Jack et al. 1995). Like some bacteriocins produced
by lactic acid bacteria (Hastings et al. 1991; Jack et al.
1995), cerein MRX1 was found both in a cell-associated
form and in the culture supernatant fluid of the producer
organism. This property of some bacteriocins to be able
to adsorb on the cellular surfaces was previously used to
design efficient adsorption desorption methods for their
recovery on a large scale. For example, the adsorption on
cells of nisin and pediocin AcH secreted in liquid media
could be promoted by adjusting the pH (of the same
media in which the strains had already grown) to 60.
The adsorption step could then be followed by cell separation and bacteriocin desorption at pH 20 (Yang et al.
1992; Daba et al. 1994). Even if inspired by those earlier
procedures, our way of cerein MRX1 recovery from the
cell surfaces was different and simpler. In fact, a relatively
large amount of cerein MRX1 was naturally adsorbed on
the cells of strain MRX1. An adsorption step by pH
adjustment was not required.
Although there are many scientific reports about the
production of bacteriocins or bacteriocin-like substances
by strains of B. cereus, purification, N-terminal amino acid
sequencing and accurate molecular mass determination
were only carried out for cerein 7A (Edman sequencing
yielded: GWGDVL), cerein 7B (Edman sequencing yielded:
GWWNSWGH) and in this work, for cerein MRX1
(N-terminus: DWTCWSCLVCAACSVELL). Both cereins
7A and 7B were isolated from B. cereus Bc7 (Oscariz et al.
2006). The determined 18-amino acid sequence of cerein
MRX1 is different from the sequences of cereins 7A and
7B but very similar to the 18-amino acid sequence of a
recently discovered 3162-Da bacteriocin, thuricin 17 produced by B. thuringiensis NEB17 (Gray et al. 2006). The
difference between the two sequences resides in the residue
in position 10 which is a cysteine in cerein MRX1 while it
is a valine in thuricin 17. A noticeable similarity between
the three cereins and thuricin 17 is the presence of one or
more tryptophan residues in their N-terminal region. Particularly, the conserved tryptophan in position 2 probably
plays an important role in the biological activity of
these four bacteriocins. Another marked similarity between
cerein MRX1 and thuricin 17 is the blockage of Edman
sequencing at the same position. This leads us to think
1629
1630
S. Sebei et al.
S. Sebei et al.
1631