Journal of Applied Microbiology 2001, 91, 636645

Purication and characterization of a bacteriocin-like

compound (Lichenin) produced anaerobically by Bacillus
licheniformis isolated from water buffalo
P. Pattnaik, J.K. Kaushik, S. Grover and V.K. Batish
Molecular Biology Unit, Dairy Microbiology Division, National Dairy Research Institute, Karnal, India
775/2/01: received 9 February 2001, revised 18 April 2001 and accepted 18 April 2001

P . P A T T N A I K , J . K . K A U S H I K , S . G R O V E R A N D V . K . B A T I S H . 2001.

Aims: To characterize a bacteriocin-like factor from Bacillus licheniformis 26 L-10/3RA

isolated from buffalo rumen.
Methods and Results: The culture supernatant exhibited the antibacterial activity against a
number of indicator organisms in a cut-well agar assay under anaerobic conditions. The
inhibitory component was puried by following ammonium sulphate precipitation, gel ltration
and ion exchange chromatography and conrmed to be a single peptide. A single band on
tricine-sodium dodecyl sulphate-polyacrylamide gel electrophoresis conrmed that the peptide
was puried to homogeneity and having an estimated molecular mass of approximately 1400
dalton. Complete amino acid sequence of the peptide yielded 12 amino acids from the
N-terminal end (ISLEICXIFHDN). No homology with previously reported bacteriocins was
observed and has been designated as Lichenin. Lichenin was found to be hydrophobic,
sensitive to atmospheric oxygen, retained biological activity even after boiling for 10 min and
was active over a pH range of 4090.
Conclusions: The Lichenin represents the rst anaerobiosis specic expression of bacteriocinlike compound isolated from Bacillus licheniformis 26 L-10/3RA of buffalo rumen origin.
Signicance and Impact of the Study: Lichenin could be a potential condidate for
manipulating the rumen function at molecular level intended for improving the productivity of
the ruminant.
INTRODUCTION
The rumen represents a complex ecosystem and an
efcient biological fermentation vessel, wherein a variety
of fermentation reactions essential for growth and productivity of ruminants are carried out by the resident
microora. The ruminal eubacterial population predominantly consists of obligatory anaerobes and few facultative
anaerobes such as Streptococcus sp., Staphylococcus sp.,
Enterococcus sp. and Lactobacillus sp. Most of these
Correspondence to: Virender K. Batish, Molecular Biology Unit, Dairy
Microbiology Division, National Dairy Research Institute, Karnal 132001,
India (e-mail: vkb@ndri.hry.nic.in).
Present address: Malaria Research Group, International Centre for Genetic
Engineering and Biotechnology, Aruna Asaf Ali Marg, PO Box 10504, New
Delhi 110 067, India.
Present address: Physical Chemistry Laboratory, Institiute for Protein
Research, Osaka University, 32 Yamadaoka, Suita 5650871, Osaka, Japan.

microora interact and compete with each other for their

survival. In a composite consortium such as the rumen,
production of antimicrobial substances such as antibiotics
and bacteriocins is a natural phenomenon employed by the
micro-organisms under an amensalistic interaction with
other inhabiting micro-organisms (Tagg et al. 1976).
Bacteriocins are a heterologous group of proteinaceous
antimicrobial substances which often display a high degree
of target specicity, although many have a wide spectrum
of activity (Jack et al. 1995). Bacteriocin production is a
common attribute in many Gram-positive bacteria. Bacteriocins and bacteriocin-like substances have been described
for a number of strict anaerobes (Barefoot and Klaenhammer 1984; Garnier and Cole 1988; Miranda et al. 1993) and
facultative anaerobes (Iverson and Millis 1976; Arihara
et al. 1993; Laukova et al. 1998a) isolated from nonruminant anaerobic sources.
2001 The Society for Applied Microbiology

CHARACTERIZATION OF LICHENIN

We have screened a number of bacterial isolates obtained

from water buffalo rumen for production of antimicrobial
substances antagonistic to Streptococcus bovis. A strain of
Bacillus licheniformis exhibited a potential antibacterial effect
against S. bovis and Eubacterium ruminantium. This culture
was also found to possess remarkable hydrolytic activities
against various polysaccharides. In this paper, we report the
isolation, purication, molecular characterization and functional activity prole of the bacteriocin-like compound
produced by B. licheniformis 26 L-10/3RA isolated from
buffalo rumen.
MATERIALS AND METHODS
Micro-organisms and culture conditions
Producer culture 26 L-10/3RA was isolated from water
buffalo rumen and identied as B. licheniformis using
morphological and biochemical tests and reconrmed by
16S rRNA sequence analysis. Other microbial cultures used as
indicator strains to generate the inhibition spectra in bacteriocin assay along with their sources are listed in Table 1. All
cultures were maintained as frozen glycerol stocks at )20C
until required. Growth medium, L-10 (Caldwell and Bryant
1966) modied for use in an anaerobic hood (Teather 1982),
consisted of glucose, maltose and cellobiose (02% w/v) as
carbon source and the solid medium contained 18% (w/v)
agar in addition. Anaerobic cultures were grown at 39C in an
anaerobic glove-box (DW Scientic, UK) maintaining an

637

atmosphere of N2-H2-CO2 gas in the ratio of 80 : 10 : 10 (v/v).

All the gases were of ultra-pure grade I. For testing under
aerobic and anaerobic conditions, L. casei ED-108 was grown
in MRS broth as well as agar. L-10 media without cystein
hydrochloride prepared aerobically was used for growth of
S. bovis for testing under aerobic conditions.
Growth conditions for production of antibacterial
factor
Flasks containing L-10 broth (pH 68) supplemented with
05% glucose (w/v) and 20% (w/v) thermocol beads were
inoculated with activated (24 h old) culture of B. licheniformis 26 L-10/3RA and incubated for 72 h at 39C. Samples
of the culture were centrifuged at 6000 g for 20 min and the
cell free culture supernatant was tested for antibacterial
activity. To prepare thermocol beads, thermocol sheets
commonly used as packaging material was separated manually into small round pieces and then placed in water and
autoclaved, leading to the compaction of the buffy round
pieces into solid small beads. The beads were then passed
through a mesh to collect uniform sized granular beads.
These thermocol beads were used to provide inert support
in the medium for the best growth of ruminal anaerobes.
Bacteriocin activity assay
Inhibitory activity in spent culture uids from B. licheniformis 26 L-10/3RA, and at each step of purication, the

Table 1 Bacterial cultures and their sources

Name of bacteria

Source

Ruminococcus albus B119

Ruminococcus avefaciens C94
Prevetolla ruminicola S-23
Selenomonas ruminantium D
Butyrivibrio brisolvens D1
Butyrivibrio brisolvens OR 3
Butyrivibrio brisolvens OR 12
Butyrivibrio brisolvens OR 515
Streptococcus bovis SB3
Streptococcus bovis 26
Eubacterium ruminantium GA-195
Fibrobacter succinogenes BL-2
Bacteroides amylophilus 70
Ruminococcus avefaciens OF-2
Ruminococcus albus A-6
Escherichia coli NCDC-136
Bacillus cereus NCDC-66
Salmonella typhi NCDC-133
Pseudomonas aeruginosa NCDC-110
Staphylococcus aureus NCDC-105
Lactobacillus casei ED-108

Bryan A. White, Department of Animal Sciences, University of Illinois, Urbana, Illinois,

Bryan A. White, Department of Animal Sciences, University of Illinois, Urbana, Illinois,
Bryan A. White, Department of Animal Sciences, University of Illinois, Urbana, Illinois,
Bryan A. White, Department of Animal Sciences, University of Illinois, Urbana, Illinois,
Ron M. Teather, Centre for Food and Animal Research, Ottawa, Ontario, Canada
Ron M. Teather, Centre for Food and Animal Research, Ottawa, Ontario, Canada
Ron M. Teather, Centre for Food and Animal Research, Ottawa, Ontario, Canada
Ron M. Teather, Centre for Food and Animal Research, Ottawa, Ontario, Canada
Ron M. Teather, Centre for Food and Animal Research, Ottawa, Ontario, Canada
Ron M. Teather, Centre for Food and Animal Research, Ottawa, Ontario, Canada
C.W. Forsberg, Department of Microbiology, University of Guelph, Canada
C.W. Forsberg, Department of Microbiology, University of Guelph, Canada
C.W. Forsberg, Department of Microbiology, University of Guelph, Canada
Molecular Biology Unit, NDRI, Karnal, India
Molecular Biology Unit, NDRI, Karnal, India
National Collection of Dairy Cultures, NDRI., Karnal, India
National Collection of Dairy Cultures, NDRI., Karnal, India
National Collection of Dairy Cultures, NDRI., Karnal, India
National Collection of Dairy Cultures, NDRI., Karnal, India
National Collection of Dairy Cultures, NDRI., Karnal, India
Multiple drug resistant milk isolate, Mol. Biol. Unit, NDRI, Karnal, India

2001 The Society for Applied Microbiology, Journal of Applied Microbiology, 91, 636645

USA
USA
USA
USA

638 P . P A T T N A I K ET AL.

lichenin was assayed by cut-well agar diffusion assay under

anaerobic conditions. Pre-poured agar media plates equilibrated under anaerobic conditions were spread with 106 cfu
of respective indicator organism and allowed to dry. In the
agar plates, wells of 5 mm diameter were cut using a cork
borer. The wells were lled with 50 ll of cell free culture
supernatant or the individual sample fractions obtained
during purication steps from the test culture and incubated overnight at 39C. The plates were then examined for
clear zones of inhibition surrounding each well and
inhibition zones were measured using a slide caliper. The
inhibitory activity was expressed in terms of arbitrary unit
(AU). Amount of antibacterial activity present in each test
sample was determined by the reciprocal of its maximum
dilution that produced a minimum of 10 mm zone of
inhibition.
Purication of Lichenin
L-10 medium (1 litre) supplemented with additional glucose
(05%, w/v) and thermocol beads (20%, w/v) was inoculated with 10 ml of fresh 24 h grown culture of
B. licheniformis 26 L-10/3RA grown in standard L-10
medium. Cultures were incubated at 39C under strict
anaerobic conditions for 72 h. Cells were then pelleted by
centrifugation (10 000 g, 20 min). The spent culture supernatant was precipitated with ammonium sulphate (80%
saturation) under chilled conditions for 12 h. The precipitate was collected by centrifugation (10 000 g, 45 min) and
the pellet was resuspended in minimum possible volumes of
10 mM phosphate buffer (pH 70) and dialysed using 1 kDa
cut-off membrane (Sigma Chemical Co.) against the same
buffer at 4C for 36 h. The insoluble material from the
dialysate was removed by centrifugation (15 000 g, 45 min)
and designated as Fraction I. The supernatant uid was
serially passed through centricon (Amicon Corp) of different
cut-off sizes beginning with 10 kDa to 3 kDa. The major
inhibitory activity was obtained in the ltrate of 10 kDa as
well as 3 kDa membrane, indicating the size of the
inhibitory factor smaller than 3 kDa.
High resolution purication of the factor was carried out
using gel ltration and ion exchange chromatographic
columns attached with FPLC system (Pharmacia). The
dialysed ltrate obtained from 3 kDa centricon passage was
applied to a gel ltration column (Superose-12, Pharmacia)
pre-equilibrated with 50 mM phosphate buffer and 150 mM
sodium chloride at pH 70 and eluted with the same buffer
at a constant ow rate. Fractions comprising 1 ml of the
eluted sample were collected manually and concentrated 10
times under vacuum. The fractions showing inhibitory
activity were pooled together and dialysed against 2 mM
Tris-HCl, pH 88 in 1 kDa dialysis membrane (Fraction II)
and concentrated 10 times under vacuum. The concentrated

sample was centrifuged at 15 000 g for 10 min to remove

any precipitated proteins. This was further loaded onto a
Mono-Q column (Pharmacia) pre-equilibrated with 20 mM
Tris-HCl, pH 88 (Solution A). The column was washed
with four column volumes of solution A to remove unbound
proteins. The bound proteins were then eluted using a
01 M linear gradient of sodium chloride (1 M NaCl +
Solution A). The 1 ml fractions collected were subjected to
cut-well agar assay for bacteriocin activity. The fractions
showing bacteriocin activity were pooled and concentrated
ve times and stored at )20C till further use (Fraction III).
The molecular weight of the puried bacteriocin
`Lichenin' was estimated by gel-ltration chromatography
(Pharmacia, Superose-12). The column was eluted with
50 mM phosphate buffer with 150 mM NaCl (pH 70) at a
constant ow rate of 05 ml min1. A standard curve
was generated using low molecular weight marker kit
(Bio-Rad).
SDS-PAGE
The purity and molecular weight of the bacteriocin-like
compound were conrmed on SDS-PAGE. The electrophoresis was carried out using a 165% polyacrylamide gel
using Tris-Tricine buffer system (Schagger and Jagov
1987). The gel was quick-stained with Coomassie brilliant
blue G-250 and destained. To localize the in situ bacteriocin
activity, the gel was cut and separated before staining and
destaining, and overlaid with 07% agar containing 106
indicator organisms. Prior to being overlaid, the gels were
xed and washed as described by Bhunia et al. (1987) and
then washed in double-distilled water. Following washing,
the gel was placed on a glass plate and allowed to
equilibrate under anaerobic conditions for 6 h (Kalmokoff
and Teather 1997) prior to overlaying with 106 organisms of
S. bovis. The overlaid gel was incubated overnight at 39C
under anaerobic conditions and examined for zones of
clearance.
Peptide analysis
Samples of puried lichenin were lyophilized under vacuum
and sequencing of the peptide was perfomed by following
sequential Edman degradation reaction using an automated
sequencer (Perkin Elmer Applied Biosystems, Model 494).
The sequence alignment to determine the homology of the
obtained sequence with the existing database at National
Centre for Biotechnology Information was carried out using
the BLAST search programme available via the website
HTTP://www.ncbi.nlm.nih.gov.
The peptide sequence of Lichenin from B. licheniformis
26 L-10/3RA has been deposited in the SwissProt databank
under accession number P82907.

2001 The Society for Applied Microbiology, Journal of Applied Microbiology, 91, 636645

CHARACTERIZATION OF LICHENIN

RESULTS

Table 2 Antibacterial spectrum of Bacillus licheniformis 26L-10/3RA

Antimicrobial spectrum and characterization

of inhibitory activity
The cell free culture supernantant (CFCS) of culture grown
in L-10 medium, pH 68 supplemented with 05% (w/v)
glucose and 20% (w/v) thermocol beads for 72 h at 39C
was tested against S. bovis SB3 and other indicator strains by
cut-well and spot agar assays (Fig. 1a,b). The antimicrobial
spectrum of bacteriocin-like substance (BLS) is given in
Table 2. As indicated therein, S. bovis SB3 was found to be
the most sensitive to the bacteriocin, lichenin produced by
B. licheniformis 26 L-10/3RA followed by S. bovis 26,
E. ruminantium GA-195, Ruminococcus albis B-199 and R.
albus A-6. The highest titre of antibacterial activity in CFCS
was found to be 2600 AU ml1 as recorded against S. bovis
SB3. The protease sensitivity of the inhibitory activity was
carried out by subjecting the CFCS to different proteolytic
enzymes (Sigma Chemical Co.) such as trypsin, a-chymotrypsin, pronase-E and proteinase-K at a concentration of
1 mg ml1. After enzymatic treatment, the samples were
incubated at 37C for 4 h. The samples were then subjected
to cut-well agar assay directed against S. bovis SB3. The
biological activity of BLS was completely inactivated by
proteinase K treatment but the same was resistant to trypsin.
This indicated the protein nature of the antimicrobial
compound that was indicated earlier as a bacteriocin-like
substance (BLS).
Production of Lichenin in liquid culture
A number of additional factors that have been reported
earlier to have a positive effect on bacteriocin production in
liquid culture in other bacterial species were also tested in
this study. Alteration in the initial pH of growth media had

(a)

639

(b)

Fig. 1 (a) Spot agar assay demonstrating antibacterial activity of

Bacillus licheniformis 26 L-10/3RA (lichenin) against Streptococcus bovis
SB3. (b) Cut-well agar assay demonstrating antibacterial activity of cell
free culture supernatant (CFCS) against Ruminococcus albus B-199

Sl.
no

Name of indicator bacteria

Sensitivity
pattern

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20

Streptococcus bovis SB3

Streptococcus bovis 26
Fibrobacter succinogenes BL-2
Bacteroides amylophilus 70
Ruminococcus avefaciens OF-2
Ruminococcus avefaciens C94
Ruminococcus albus B119
Ruminococcus albus A-6
Butyrivibrio brisolvens D1
Butyrivibrio brisolvens OR 3
Butyrivibrio brisolvens OR 12
Butyrivibrio brisolvens OR 515
Selenomonas ruminantium D
Eubacterium ruminantium GA-195
Lactobacillus casei ED-108
Escherichia coli NCDC-136
Bacillus cereus NCDC-66
Pseudomonas aeruginosa NCDC-110
Staphylococcus aureus NCDC-105
Salmonella typhi NCDC-133

+
+
)
)
+
+
+
+
)
)
+
)
)
+
+
)
)
)
)
)

++
++

+
+
++
++

++

Zone
diameter
(mm)*
35
33
)
)
15
13
28
28
)
)
8
)
)
31
7
)
)
)
)

) Absence of inhibition. + Zone of inhibition of 712 mm

diameter, ++ Zone of inhibition of 1217 mm diameter, +++ Zone
of inhibition of 1721 mm diameter, *Well size 50 mm diameter.

no effect on the induction of inhibitory activity. Supplementation of glucose at a level of 05% (w/v) to the growth
medium exerted a stimulatory effect on Lichenin production; however, increasing glucose beyond the level of 10%
(w/v) caused a drastic reduction in BLS activity. L-10
medium is a semidened medium and contains no rumen
uid. The medium was designed for growth and enumeration of a wide variety of ruminal isolates (Caldwell and
Bryant 1966). Inclusion of solid phase in the form of straw
powder and sawdust up to 1% (w/v) to the culture medium
caused only a marginal increase in Lichenin production.
However, signicant improvements in nal culture density
resulting from inclusion of 20% (w/v) thermocol beads
yielded high level of detectable protease sensitive inhibitory
activity in CFCS from B. licheniformis 26 L-10/3RA.
Lichenin production by B. licheniformis 26 L-10/3RA was
observed only under anaerobic conditions and the antibacterial activity was also demonstrated against S. bovis SB3
only under anaerobic conditions. Exposure of the culture
supernatant to the atmosphere resulted in the loss of
inhibitory activity, but when the CFCS was replaced under
anaerobic conditions it regained the inhibitory activity. We
also observed a correlation between growth of B. licheniformis 26 L-10/3RA and production of Lichenin. Antibacterial
activity appeared in culture supernatant from the late

2001 The Society for Applied Microbiology, Journal of Applied Microbiology, 91, 636645

640 P . P A T T N A I K ET AL.

logarithmic growth phase to early stationary stage and

reached a titre of 2000 AU within the next 18 h. Production
continued slowly for another 24 h up to a maximum titre of
2600 AU.
Several bacteriocins are known to be cell associated
(Kalmokoff et al. 1996); however, we obtained only 20% of
the activity associated with cell and the majority of it was
present in CFCS. Addition of Tween-20 to the culture
broth did not have any appreciable effect on the release of
the bound Lichenin. On the other hand, Tween-20 was
itself inhibitory to the growth of S. bovis SB3.

process of anion-exchange chromatography the antibacterial

fraction became eluted as a single peak in the washing step
before the application of NaCl gradient (Fig. 2b). It was
puried up to homogeneity and 22523-fold of purication
was achieved, but the nal recovery/yield was only 60%.
Tricine-SDS-polyacrylamide gel electrophoresis followed
by Coomassie blue G-250 staining indicated that the peak
consisted of a single peptide with an estimated molecular
mass of 1400 dalton (Fig. 3a). The localization of in situ
inhibitory activity against S. bovis SB3 (Fig. 3b) conrmed
the band responsible for antibacterial action.

Purication of Lichenin

Effect of pH and heat treatment

The inhibitory antibacterial component was isolated from

CFCS by a combination of ammonium sulphate precipitation, gel ltration and ion exchange chromatography and the
results have been presented in Table 3. Gel ltration
resulted in fractions exhibiting antibacterial activity corresponding to three poorly resolved peaks. A repeat run of the
pooled fractions showing activity did not result in much
improvement in the resolution of peaks (Fig. 2a). Since
antibacterial activity is present over a wide range of elution
volume (160235 ml.) and given the fact that proteins in
these fractions are not well resolved, it was difcult to
determine precisely the elution volume for proteins having
antibacterial activity. However, as the maximum zone of
inhibition (21 cm) was observed at 195 ml of elution, this
point was considered arbitrarily for determination of
molecular weight of the antibacterial protein. The approximate molecular weight of the BLS or inhibitory compound
as per the theoretical calculation was found to be 1478
dalton.
As a preliminary trial, a mapping study was performed to
determine the exchanger mode and corresponding optimal
pH for the best separation of BLS. In the pH mapping
study, Q-sepharose (fast ow) as an anion and CMsepharose (fast ow) as a cation exchanger were employed
in a pH range from 40 to 90. The inhibitory component did
not bind to any of the matrix at all studied pH values. In the

Puried Lichenin retained biological activity at all the pH

values ranging from 40 to 90. It was stable under
pasteurization treatment and boiling at pH 70. However,
heat treatment at 80C and boiling for 10 min at pH 40 and
90 resulted in a signicant reduction of biological activity
(data not shown). The puried peptide lost 385% activity
upon treatment with a-chymotrypsin and pronase E, and a
complete loss of antibacterial activity (100%) upon proteinase K treatment. There was no loss of antibacterial activity
when stored for up to 6 days at room temperature.
However, a total loss in activity was observed on day 9 of
storage. At refrigerated temperature, it exhibited full activity
up to day 16 of storage followed by a slow and gradual
decrease in biological activity.
Amino acid sequence analysis
The amino acid sequencing of the active peptide indicated
the presence of a total of 12 amino acids in the peptide.
Calculation of molecular weight from the peptide sequence
resulted an approximate molecular weight of 1344 kDa that
is consistent with the molecular weight of the peptide,
1478 kDa obtained using gel ltration chromatography.
The amino acid sequence obtained was ISLEICXIFHDN;
no N-terminal block was observed in the sequence and the
peptide did not show any characteristics of cyclicity.

Table 3 Purication index of BLS produced by Bacillus licheniformis 26L-10/3RA

Fractions

Volume
(ml)

Total
protein (lg)

Total
activity (AU)

CFCS
Fraction I
Fraction II
Fraction III

1000
24
8
1

19403
1813
560
55

26
13
853
166

106
106
105
105

Specic activity
(AU l)1g)

Relative specic activity

(purication fold)

Yield (activity
recovered) (%)

134
717
1489
30181

1
535
1111
22523

100
50
32
6

Total protein protein (lg) total volume (ml). Total activity activity (AU) total volume (ml). Specic activity total activity
(AU)/total protein (lg). Relative specic activity specic activity of same sample/Initial specic activity. Yield [total activity of the
sample/Initial total activity] 100.
2001 The Society for Applied Microbiology, Journal of Applied Microbiology, 91, 636645

CHARACTERIZATION OF LICHENIN

641

Fig. 2 (a) Elution prole of Lichenin from gel-ltration column

(Superose-12). (b) Elution prole of Lichenin from ion-exchange
column (Mono-Q). The area shaded corresponds to fractions showing
bacteriocin-like activity

(a)

(b)

Fig. 3 (a) Tricine-SDS-PAGE. 1. Molecular weight marker. 2.

Puried Lichenin. (b) Direct overlay of the neutralized SDS-PAGE gel
demonstrating clear inhibition zone against Streptococcus bovis SB3
around the lichenin band

However, the seventh amino acid residue could not be

identied and it did not belong to any of the natural amino
acids. On comparison with PIR/PDB database, no homology with any previously reported bacteriocin or other
proteins sequences was found. Therefore, this unique
peptide has been named as `Lichenin' and the sequence is
submitted to SwissProt database with an accession number
P82907.
Co-culture studies
The co-culture studies of the producer B. licheniformis
26 L-10/3RA and S. bovis SB3 could not yield any
conclusive result due to the wide variation in their
respective growth pattern. However, Lichenin exhibited a
marked bactericidal effect on exponential phase culture of
S. bovis SB3 (Fig. 4). Upon addition of partially puried
preparation of Lichenin (150 AU ml1) the number of
viable cells decreased considerably during the rst 2 h and
the turbidity of the culture remained fairly constant over
the rest of the incubation period. The reduction in cell
2001 The Society for Applied Microbiology, Journal of Applied Microbiology, 91, 636645

642 P . P A T T N A I K ET AL.

1012
1011
1010
109
108
107
106
105
104
103
102
0

20

40

60

80

Fig. 4 Antibacterial effect of partially puried lichenin against

exponential culture of Streptococcus bovis SB3. Concentration of BLS:
j, Control; d, AU50; m, AU100; ., AU150.

numbers caused by addition of 150 AU ml1 of partially

puried lichenin was more pronounced and drastic than
was caused by the addition of 100 AU ml1 Lichenin
(results not shown).
DISCUSSION
As has been observed with many other secondary metabolites, B. licheniformis 26 L-10/3RA also produced antibacterial factor during the early stationary growth phase.
Initially, inhibitory activity was observed only on solid agar
medium with spot agar assay (Fig. 1a) and cut-well assay
failed to detect inhibitory activity using CFCS from up to
48-h-old culture. It indicated that the BLS might possibly
be partially or fully cell/membrane bound. Several other
workers (Nakamura et al. 1981; Southern et al. 1984) have
also reported production of cell-bound bacteriocins or
bacteriocin-like substances by facultative anaerobic bacteria.
Several of the enzyme complexes in rumen bacteria are
known to be cell-bound; our results also indicated the
similar pattern of production of inhibitory factor and, hence,
showed antagonistic activity only on solid substrates (agar
media). Contact-mediated regulation of bacteriocin production could be an effective economy for the producing cell,
particularly in the rumen, where competition for energy
involves concurrent competition for attachment sites (Kalmokoff et al. 1996). Sonication of the cells recovered from
48-h grown culture yielded a detectable level of antibacterial
activity in the lysate and the major fraction of inhibitory
activity remained conned to the cell debris. However,
incubation up to 96 h also yielded detectable levels of

antibacterial activity in culture broth with a minor fraction

still remaining within the cell.
The production and biological activity of Lichenin has
been found to be strictly dependent upon the absence of
oxygen. Our results are in strong agreement with Kalmokoff
and Teather (1997), reporting the loss of inhibitory activity
of the antibacterial compound `butyrivibriocin' by Butyrivibrio brosolvens AR10, when extracts of the bacteriocin
preparations were exposed to the atmosphere. Bacteria often
encounter a drastic change in their environment, including
uctuation in the levels of external oxygen. Unlike obligate
aerobes or obligate anaerobes which can survive only either
in the presence or absence of oxygen, facultative anaerobes
can adjust with changes in environmental oxygen level by
sensing oxygen concentration and shifting cellular metabolism accordingly (Gunsalus and Park 1994; Luchi and
Weiner 1996; Sawer 1999). The changes in metabolism in
response to changes in oxygen availability include adjustment with the rate and route of carbon source utilization, the
pathway of electron ow to maintain an oxidation-reduction
balance, and the mechanism of energy production and
biosynthetic reactions. These changes are achieved by
modulating protein activity or by regulating the expression
of the appropriate genes, or both. Specic gene expression in
Escherichia coli under aerobic and anaerobic conditions are
now well established (Smith and Neidhardt 1983a,1983b).
Nakano and Zuber (1988) reported that most but not all of
the aerobically induced genes were also induced in anaerobic
conditions in the presence of nitrate. However, the majority
of anaerobically induced genes were repressed in the
presence of alternative electron acceptors under aerobic
conditions. B. subtilis, a strict aerobe, when shifted to
anaerobic conditions undergoes a physiological reorientation
leading to expression of a cascade of anaerobiosis-specic
genes (res, nar and fnr gene cluster) whose gene products
help the organism to adapt and survive successfully in
oxygen-limited conditions. However, these genes fail to
express under aerobic growth conditions (Nakano and Zuber
1998). We assume the existence of a similar pattern of
regulation of bacteriocin by our producer strain B. licheniformis 26 L-10/3RA. The anaerobiosis-specic antibacterial
activity observed in our isolate may be attributed to
anaerobiosis specic expression.
In this study, CFCS obtained from anaerobic growth of
producer strain did not show any inhibition against the
indicator strain S. bovis SB3 under aerobic conditions. This
may be attributed to the loss of antibacterial activity of BLS
when exposed to air or development of resistance mechanism by indicator strain when grown aerobically. However, to
date there is no published report available concerning
development of resistance to any antimicrobial agents by
S. bovis (sensitive to these agents under anaerobic conditions) when grown aerobically. These experiments could not

2001 The Society for Applied Microbiology, Journal of Applied Microbiology, 91, 636645

CHARACTERIZATION OF LICHENIN

be conducted with any other sensitive indicator bacteria of

ruminal origin as they failed to grow under aerobic
conditions, thereby limiting the credibility of inhibitory
activity under anaerobic conditions only to facultative
anaerobes such as S. bovis SB3. Most of the aerobic
organisms possess a strong oxidative status of the cell
membrane which helps them in oxidative respiration,
whereas this is not so in anaerobes, as oxygen does not
serve as the terminal electron acceptor. S. bovis SB3 grown
aerobically may have an high oxidative status of the cell
membrane which causes inactivation of Lichenin when
internalized by cells during aerobic growth, whereas the
same may be failing to do so when grown anaerobically
because of low redox potential. CFCS obtained from aerobic
growth of the producer strain also did not show any
inhibition of S. bovis SB3 in either the aerobic or anaerobic
conditions, which may be attributed to some sort of
inexplicable repression in the expression of the genes
encoding Lichenin at aerobic conditions.
Production of BLS by the test culture was invariably
observed only in small volumes (10 ml) and it was adversely
affected when scaled up to 100 ml. On the other hand,
production was observed consistently on solid medium,
suggesting that the production of BLS by B. licheniformis
26 L-10/3RA might require the intervention of solid
support to mimic the rumen environment which itself is
semisolid with a pack of brous feed particles. Addition of
thermocol beads at a level of 20% (w/v) in the growth
medium led to a substantial increase in the production of
Lichenin. A sharp increase in the production of Lichenin by
addition of inert thermocol beads may be attributed to
availability of increased surface area for attachment and
subsequent growth of the producer strain or contact
regulation, i.e. induction of Lichenin by solid substrates.
However, it was reported earlier that inclusion of a solid
phase to the media did not have any effect on induction of
inhibitory activity of Butyrivibrio brisolvens AR10 (Kalmokoff and Teather 1997). Surprisingly, addition of straw
powder and saw dust in the growth medium caused only a
marginal increase in BLS production. This anomaly could
be attributed to strong cellulolytic and xylanolytic activity of
the producer strain leading to degradation/utilization of
these substrates and hence their disappearance from the
medium.
We observed frothy occulation and development of
pelicular layer upon overnight storage of CFCS at 80%
saturation of ammonium sulphate at refrigeration temperature. Subsequent to ammonium sulphate precipitation, we
failed to redissolve the precipitate completely in aqueous
buffer. This indicated that the precipitate containing
antibacterial fraction may be hydrophobic in nature. A good
amount of precipitate settled after dissolution in buffer,
hence the insoluble material was discarded by high-speed

643

centrifugation to avoid interference in the subsequent

purication steps, which may be one of the reasons for
poor recovery of the nal puried peptide.
As the antibacterial component did not bind to any of the
ion-exchanger matrix, Mono-Q anion exchanger was selected so that contaminating proteins could bind strongly and
be retained in the column while the antibacterial activity
could be eluted in the washings. Silver staining failed to
detect any peptide band on the Tricine-SDS-polyacrylamide gel. However, staining with Coomassie blue G-250,
followed by a quick destaining process, resulted in a clearly
visible single band. Methanol from the destaining solution
was omitted and the process was carried out at 65C to avoid
diffusion of protein during destaining process. Nevertheless,
we could not obtain a sharp band, which may be due to the
gradual diffusion of the peptide after the applied voltage
across the gel during electrophoresis was withdrawn.
Amino acid residues at all the positions were conrmed,
except at position 7. Presence of cystein residue at the sixth
position from the N-terminal end of BLS may explain why
no antibacterial activity was observed under aerobic conditions. In Lichenin, oxidation or reduction status of cystein
residues may be playing an important role in conferring
antibacterial activity to the whole peptide. The amino acid
residue at the seventh position could not be identied, as it
resulted in a very large peak which did not correspond to
any of the currently known amino acids. This residue may
be some unnatural or modied amino acid. Bacteriocins or
antibacterial peptides such as nisin (Gross and Morell 1971),
subtilin (Gross and Kiltz 1973), epidermin (Allgaier et al.
1985), Pep5 (Kellner et al. 1989), gallidermin (Kellner et al.
1988), etc. also contain unusual amino acids such as
lanthionine. Incorporation of a lanthionine residue introduces a monosulphur bridge, which results in a unique peptide
ring structure, especially in antibiotics. However, in our
opinion the unidentied seventh residue of the BLS may not
be a lanthionine residue, because such ring-forming amino
acids result in blank cycles with the Edman degradation
(Kellner et al. 1988; Kellner et al. 1989) and no such blank
cycle was observed in the sequencing. The N-terminal
sequence analysis of BLS indicates the presence of a higher
number of hydrophobic amino acids in the peptide, which
may also be causing the occulation observed upon precipitation with ammonium sulphate and afterwards incomplete dissolution of ammonium sulphate precipitated protein
in aqueous buffer.
Many antimicrobial peptides are hydrophobic in nature,
which promotes interaction with cell membranes. Nisin,
Pep5 and subtilin are known to form pores in the bacterial
membranes because of their strong hydrophobic nature
(Kordel et al. 1989; Schuller et al. 1989). The broadspectrum inhibitory activity of Lichenin against a wide
range of rumen bacteria (Fig. 1b) could be due to its

2001 The Society for Applied Microbiology, Journal of Applied Microbiology, 91, 636645

644 P . P A T T N A I K ET AL.

amphipathic nature (as it contains both hydrophobic and

hydrophilic residues) which could cause it to have surfactant-like activity on cell membrane, thereby disrupting
cellular function. However, more in-depth study is required
to explain the mode of action of Lichenin.
Purication and amino acid sequence analysis of this
novel peptide will help us in further understanding of
bacteriocins and other antibacterial peptides produced
anaerobically by many obligatory and facultative anaerobic
bacteria. Purication of bacteriocin and bacteriocin-like
substances produced by anaerobic bacteria isolated from
habitats such as the rumen has been a very difcult task
due to several factors, especially the tendency of such
molecules to lose activity in the presence of oxygen. The
hydrophobicity exhibited by numerous antimicrobial peptides, including Lichenin, could be a general characteristic
of such peptides required for interactions with membranes.
The detailed elucidation of Lichenin structure will provide
understanding of the proteinmembrane interactions at
molecular levels and elucidation of the amino acid sequence
of Lichenin would facilitate cloning of the gene for this
antibacterial peptide. Further characterization of Lichenin
and its genetic determinants may show this small peptide
to be a model for studying anaerobiosis-specic expression
of antibacterial proteins and bacteriocins for studying
bacteriocin structurefunction relationships, hostrange
interaction and the physiology of bacteriocin production
and immunity among the obligatory and facultative
anaerobic bacteria.
ACKNOWLEDGEMENTS
We thank Dr C.A. Reddy (Michigan State University, East
Lansing, USA) for his assistance in amino acid sequencing
of the peptide and Dr O.P. Kupers, NIZO, the Netherlands
for his help in 16S rRNA analysis. The nancial assistance
offered by Council for Scientic and Industrial Research
(CSIR), India to the rst author in the form of Senior
Research Fellowship is duly acknowledged.
REFERENCES
Allgaier, H., Jung, G., Werner, R.G., Schneider, U. and Zahner, H.
(1985) Elucidation of the structure of epidermin, a ribisomally
synthesized tetracyclic heterodetic polypeptide antibiotic. Angew
Chemica 24, 10511053.
Arihara, K., Cassens, R.G. and Luchansky, J.B. (1993) Characterization of bacteriocins from Enterococcus faecalis with activity against
Listeria monocytogenes. International Journal of Food Microbiology 19,
123134.
Barefoot, S.F. and Klaenhammer, T.R. (1984) Purication and
characterization of the Lactobacillus acidophilus bacteriocin Lactacin
B. Antimicrobial Agent Chemotherapy 26, 328334.

Bhunia, A.K.M., Johnson, M.C. and Ray, B. (1987) Direct detection of

an antimicrobial peptide of Pediococcus acidilactici in sodium dodecyl
sulphate polyacrylamide gel electrophoresis. Journal of Industrial
Microbiology 2, 319322.
Caldwell, D.R. and Bryant, M.P. (1966) Medium without rumen uid
for non-selective enumeration and isolation of rumen bacteria.
Applied Microbiology 14, 794801.
Garnier, T. and Cole, S.T. (1988) Complete nucleotide sequence and
genetic organization of the bacteriocinogenic plasmid, pIP404, from
Clostridium perfringens. Plasmid 19, 134150.
Gross, E. and Kiltz, H. (1973) The number and nature of a,
b-unsaturated amino acids in subtilin. Biochemistry Biophysics
Research Communication 50, 559565.
Gross, E. and Morell, J.L. (1971) The structure of nisin. Journal of
American Chemical Society 93, 46344635.
Gunsalus, R.P. and Park, S.J. (1994) Aerobicanaerobic gene regulation in Escherichia coli: control by ArcAB and Fnr regulons. Research
Microbiology 145, 437450.
Iverson, G.W. and Millis, N.F. (1976) Bacteriocin of Streptococcus
bovis. Canadian Journal of Microbiology 22, 10401047.
Jack, R.W., Tagg, J.R. and Ray, B. (1995) Bacteriocins of Grampositive bacteria. Microbiology Review 59, 171200.
Kalmokoff, M.L., Bartlett, F. and Teather, R.M. (1996) Are rumen
bacteria armed with bacteriocins? Journal of Dairy Science 79,
22972306.
Kalmokoff, M.L. and Teather, R.M. (1997) Isolation and characterization of a bacteriocin Butyrivibriocin AR10 from the ruminal
anaerobe Butyrivibrio brisolvens AR10: evidence in support of wide
spread occurrence of bacteriocin-like activity among ruminal isolates
of B. brisolvens. Applied Environmental Microbiology 63, 394402.
Kellner, R., Jung, G., Horner, T., Zahner, H., Schnell, N., Entian,
K.D. and Gutz, F. (1988) Gallidermin, a new lanthionine-containing
polypeptide antibiotic. European Journal of Biochemistry 177, 5359.
Kellner, R., Jung, G., Josten, M., Kaletta, C., Entian, K.D. and Sahl,
H.G. (1989) Pep5: structure elucidation of a large lantibiotic. Angew
Chemistry International 28, 61619.
Kordel, M., Schuller, F. and Sahl, H.G. (1989) Interaction of the pore
forming-peptide antibiotics Pep5, nisin and subtilin with nonenergized liposomes. FEBS Letters 244, 99102.
Laukova, A., Czikkova, S., Vasilkova, Z., Juris, P. and Marekova, M.
(1998a) Occurrence of bacteriocin production environmental enterococci. Letters in Applied Microbiology 27, 178182.
Luchi, S. and Weiner, L. (1996) Cellular and molecular physiology of
Escherichia coli in the adaptation to aerobic environments. Journal of
Biochemistry 120, 10551063.
Miranda, C.M.S., Farias, L.M., Carvalho, A.R., Damasceno, C.A.V.,
Totola, A.H., Tavares, C.A.P., Cisalpino, E.O. and Vieira, E.C.
(1993) Purication and partial characterization of a bacteriocin
isolated from Bacteroides ovatus H47. Canadian Journal of Microbiology 39, 169174.
Nakamura, T., Fujimura, S., Obata, N. and Yamazaki, N. (1981)
Bacteriocin-like substance melaninocin. from oral Bacteroides melanogenicus. Infection and Immunity 31, 2832.
Nakano, M.M. and Zuber, P. (1998) Anaerobic growth of a `strict
aerobe' Bacillis subtilis. Annual Reviews in Microbiology 52, 165190.
Sawer, G. (1999) The aerobic/anaerobic interface. Current Opinion in
Microbiology 2, 181187.

2001 The Society for Applied Microbiology, Journal of Applied Microbiology, 91, 636645

CHARACTERIZATION OF LICHENIN

Schagger, H. and Jagow, G.V. (1987) Tricine-sodium dodecyl

sulphate-polyacrylamide gel electrophoresis for the separation of
proteins in the range from 1 to 100 kDa. Analytical Biochemistry 166,
368379.
Schuller, F., Benz, R. and Sahl, H.G. (1989) The peptide antibiotic
subtilin acts by formation of voltage-dependent multi-state pores in
bacterial and articial membranes. European Journal of Biological
Chemistry 182, 181186.
Smith, M.W. and Neidhardt, F.C. (1983a) Proteins induced by
anaerobiosis in Escherichia coli. Journal of Bacteriology 154, 336343.

645

Smith, M.W. and Neidhardt, F.C. (1983b) Proteins induced by

aerobiosis in Escherichia coli. Journal of Bacteriology 154, 344350.
Southern, J.A., Katz, W. and Woods, D.R. (1984) Purication and
properties of a cell-bound bacteriocin from a Bacteroides fragilis
strain. Antimicrobial Agent Chemotherapy 25, 253257.
Tagg, J.R., Dajani, A.S. and Wannamaker, L.W. (1976) Bacteriocins of
Gram-positive bacteria. Bacteriology Reviews 40, 722756.
Teather, R.M. (1982) Isolation of plasmid DNA from
Butyrivibrio brisolvens. Applied Environmental Microbiology 43,
298302.

2001 The Society for Applied Microbiology, Journal of Applied Microbiology, 91, 636645