Biochimica et Biophysica Acta 1770 (2007) 495 – 505
www.elsevier.com/locate/bbagen
Low molecular weight chitosans—Preparation with the
aid of pronase, characterization and their bactericidal
activity towards Bacillus cereus and Escherichia coli
Acharya B. Vishu Kumar a , Mandyam C. Varadaraj b ,
Lalitha R. Gowda c , Rudrapatnam N. Tharanathan a,⁎
a
Department of Biochemistry and Nutrition, Central Food Technological Research Institute, Mysore-570020, India
b
Human Resource Development, Central Food Technological Research Institute, Mysore-570020, India
c
Protein Chemistry and Technology, Central Food Technological Research Institute, Mysore-570020, India
Received 3 March 2006; received in revised form 8 December 2006; accepted 12 December 2006
Available online 20 December 2006
Abstract
The homogeneous low molecular weight chitosans (LMWC) of molecular weight 9.5–8.5 kDa, obtained by pronase catalyzed non-specific
depolymerization (at pH 3.5, 37 °C) of chitosan showed lyses of Bacillus cereus and Escherichia coli more efficiently (100%) than native
chitosan (< 50%). IR and 1H-NMR data showed decrease in the degree of acetylation (14–19%) in LMWC compared to native chitosan (∼26%).
Minimum inhibitory concentration of LMWC towards 106 CFU ml− 1 of B. cereus was 0.01% (w/v) compared to 0.03% for 104 CFU ml− 1 of E.
coli. SEM revealed pore formation as well as permeabilization of the bacterial cells, as also evidenced by increased carbohydrate and protein
contents as well as the cytoplasmic enzymes in the cell-free supernatants. N-terminal sequence analyses of the released proteins revealed them to
be cytoplasmic/membrane proteins. Upon GLC, the supernatant showed characteristic fatty acid profiles in E. coli, thus subscribing to detachment
of lipopolysaccharides into the medium, whereas that of B. cereus indicated release of surface lipids. The mechanism for the observed bactericidal
activity of LMWC towards both Gram-positive and Gram-negative bacteria has been discussed.
© 2006 Elsevier B.V. All rights reserved.
Keywords: Low molecular weight chitosan; Pronase; Structure; Bactericidal activity; Mechanism
1. Introduction
Chitin, next to cellulose is the most abundant natural aminopolysaccharide on Earth. Commercially it is found in the offal of
marine food processing industry [1], and as only a small
quantity of the offal is utilized for animal feed, its disposal is of
environmental concern. In the past two decades, chitosan, a β
1 → 4 copolymer of glucosamine (GlcN) and N-acetylglucosamine (GlcNAc) residues obtained by partial de-N-acetylation of
Abbreviations: Mr, Molecular weight; DA, Degree of acetylation; DP,
Degree of polymerization; CFU, Colony forming units; MIC, Minimum
inhibitory concentration; HPSEC, High performance size exclusion chromatography; SEM, Scanning electron microscopy; GPC, Gel permeation chromatography; GLC, Gas–liquid chromatography; LPS, lipopolysaccharides
⁎ Corresponding author. Tel.: +91 821 2512685; fax: +91 821 2517233.
E-mail address: [email protected] (R.N. Tharanathan).
0304-4165/$ - see front matter © 2006 Elsevier B.V. All rights reserved.
doi:10.1016/j.bbagen.2006.12.003
chitin, drew much attention owing to its better solubility and
reactivity compared to chitin, and thus exhibiting enhanced biofunctionalities such as antimicrobial, antitumor, hypolipidemic,
hypocholesterolemic, and immuno-stimulating activities [2,3].
Recent studies have revealed the antimicrobial potential of
chitosan to be dependent on its Mr as well as DA [4,5].
However, high Mr and high viscosity of chitosan solutions have
restricted its multidimensional utility. On the other hand, the
LMWC obtained by physical, acidic or enzymatic depolymerization of chitosan, due to its ready solubility in water, is better
amenable for a wide variety of biomedical applications. It was
reported that LMWC (5–10 kDa) had highest bactericidal
activity towards pathogenic bacteria [5], whereas a 20 kDa
product prevented progression of Diabetes mellitus and showed
a higher affinity for lipopolysaccharides (LPS) than the native
chitosan of ∼ 140 kDa [6]. The practical use of LMWC in milk
496
A.B. Vishu Kumar et al. / Biochimica et Biophysica Acta 1770 (2007) 495–505
preservation and oral hygiene is also reported [7]. Of late,
LMWC of 5–10 kDa was also shown to have potential as DNA
delivery system [8].
Chitosan could be depolymerized by physical, chemical or
enzymatic methods [9–11]. The former (e.g., sonication,
shearing, etc.) requires special equipments and the chemical
hydrolysis using HCl, H2SO4, H2O2, HNO2 do not lend
themselves to easy reaction control and often results in the
modification of the depolymerization products. Enzymatic
depolymerization based either on specific or non-specific
enzymes finds advantages over other methods, as it overcomes
the cited drawbacks and also the reaction will be under facile
control. Chitosanase, the specific enzyme for chitosanolysis is
expensive, unavailable in bulk and results in a preferential
formation of chitooligomers-monomers due to its specificity,
whereas the non-specific enzymes, which are inexpensive and
commercially available, result mainly in the formation of
LMWC [12,13]. By varying the reaction conditions such as pH
of the reaction medium, temperature and time, it was possible to
obtain LMWC of Mr in the range 9.0 ± 0.5 kDa by depolymerizing chitosan using pronase, a non-specific enzyme [14]. In the
present study, further characterization of LMWC and the effect
of its Mr and DA on bactericidal activity towards Gram-positive
(Bacillus cereus) and Gram-negative (Escherichia coli) bacteria
as well as the mechanism of their action are reported.
2. Experimental procedures
2.1. Materials
Chitosan standards (Mr, 150–600 kDa) were obtained from Fluka Chemical
Corp., Switzerland. Chitosan 10 (Mr, 60 kDa) was from Wako Chemical, Osaka,
Japan. Pronase (Type XXV protease from Streptomyces griseus, EC. 3.4.24.4)
was from Sigma Chemical Co., St. Louis, MO, USA. Dextran standards of
Mr 10–70 kDa were from Pharmacia Fine Chemicals, Uppsala, Sweden and
1–5 kDa were from Fluka Chemika, USA. Shrimp chitin was from CFTRI
Regional Center at Mangalore, India. Other chemicals used were of highest
purity available.
2.2. Preparation of chitosan
Shrimp chitin was subjected to heterogeneous N-deacetylation to obtain
chitosan [15], which was further purified by dissolving in 1% acetic acid, filtered
and precipitated with 2% sodium carbonate. The precipitate was water washed
and freeze-dried (native chitosan).
2.3. Isolation of LMWC
Chitosan solution (1%, in 1% acetic acid and pH adjusted to 3.5 with 0.1 N
HCl/NaOH) was treated with pronase in the ratio 100:1 (w/w), incubated for
different periods (1, 3 and 5 h) at 37 °C followed by arresting the reaction by
heat denaturing the enzyme (100 °C, 5 min) and adding equal volume of 2 N
NaOH. The precipitate (LMWC) obtained after centrifugation (3000 rpm,
10 min) was dialyzed against deionized water using a membrane (Sigma
Chemicals Co, USA) having Mr cut-off 2 kDa at ambient temperature and
freeze-dried.
Houwink's equation, (η) = KMar where (η) = intrinsic viscosity, K = 3.5 × 104,
a = 0.76 [16].
2.5. GPC and HPSEC
The Mr of chitosan was determined by GPC on a Sepharose CL-4B column
(Sigma, bed volume—180 ml) [4]. LMWC was analyzed by HPSEC on
E-linear-E-1000 columns (Waters Associates, Milford, Massachusetts, USA)
connected in series to a RI detector (Shimadzu LC-8A system) and GPC on
Biogel P30 (Bio Rad laboratories, CA, Bed volume—100 ml). Acetate buffer
(pH 4.5) was used as the eluant and both the columns were pre-calibrated with
dextran and chitosan Mr standards.
2.6. IR spectroscopy
IR spectral studies were performed in a Perkin Elmer spectrum 2000
spectrometer (Connecticut, USA) under dry air at room temperature using KBr
pellets. Chitosan and LMWC (4 mg each) were mixed thoroughly with 200 mg
KBr; 40 mg of the mixture was palletized and subjected to IR spectroscopy.
Reproducibility of the spectra was verified on three preparations and DA was
1
−1
determined using the formula, (A−1655
cm / A3450 cm) × 100 ÷ 1.33, where A—
absorbance at these wavelengths, calculated from baseline drawing [17].
2.7. Liquid state 1H-NMR
Native chitosan and LMWC (50 mg each) dissolved in 1 ml solvent mixture
of D2O + DCl (0.98 + 0.02 ml, respectively) were subjected to 1H-NMR on a
Jeol 300 MHz spectrometer at 23 °C. Degree of deacetylation (DDA, %) was
calculated using integrals of the peaks of proton H1 of deacetylated monomer
(H1D) and the three protons of acetyl group (H-Ac), i.e., DDA (%) = H1D /
(H1D + H-Ac/3) × 100 [18].
2.8. Chitosan/LMWC—solubility study
Solubility of chitosan/LMWC was determined according to the method of
Qin et al. with slight modification [19]. Sample (chitosan/LMWC, 0.1 g) was
suspended in 10 ml solvent (distilled water, 0.01% and 1% acetic acid) at 25 °C
for 2 h under constant stirring. Soluble chitosan/LMWC was removed by
centrifugation at 5000 rpm for 15 min. The precipitate was washed thoroughly
with ethanol, collected and weighed after drying over phosphorous pentoxide
in vacuum. The solubility of chitosan/LMWC (average of five trials) was
determined by the percent of chitosan/LMWC dissolved.
2.9. Indicator bacteria and inoculum preparation
Strains of Bacillus cereus F4810 (courtesy, Dr. J.M. Kramer, Central Public
Health Laboratory, United Kingdom), Escherichia coli D21 (courtesy, Dr. M.A.
Linggood, Unilever Research, United Kingdom), Listeria monocytogenes
Scott A (courtesy, Dr. Arun K. Bhunia, Purdue University, USA), Yersinia
enterocolitica MTCC 859 (courtesy, Microbial Type Culture Collection,
Institute of Microbial Technology, Chandigarh, India), Staphylococcus aureus
FRI 722 (courtesy, Dr. E. Notermans, Public Health Laboratory, The Netherlands) and Bacillus licheniformis CFR 1621 (Native food isolate) were obtained
from the culture collection maintained in the Institute.
The cultures were maintained at 6 °C on brain heart infusion (BHI) agar
(HiMedia, Mumbai, India) slants and sub-cultured at 15-day intervals. Prior to
use, the culture was successively propagated twice in BHI broth at 37 °C. Cell
suspensions of the culture, individually, were prepared from 20 h old BHI
culture broth with appropriate dilution in 0.85% saline, giving individual counts
of 102–106 CFU ml− 1 [20].
2.10. Bacterial growth inhibitory assay
2.4. Determination of the Mr
Viscometric measurement: The viscosity of chitosan dissolved in sodium
acetate buffer (0.5 M acetic acid + 0.2 M sodium acetate, pH 4.5) was measured
using an Ostwald's viscometer. The average Mr was deduced using the Mark–
To 9 ml aliquots of nutrient broth (HiMedia, Mumbai, India) containing
0.5% dextrose, added 1 ml of 0.1% chitosan and LMWC dissolved in 0.1 and
1% acetic acid, respectively and the broth pH was adjusted to 6.0 with sodium
acetate, so as to get final chitosan/LMWC concentrations between 0.01–0.1%
A.B. Vishu Kumar et al. / Biochimica et Biophysica Acta 1770 (2007) 495–505
497
(1–10 mg/10 ml broth solution). Nutrient broths containing 0.01 and 0.1% acetic
acid and pH adjusted to 6.0 served as controls. To the broth, added a cell
suspension (100 μl) of specific bacterial strain in the concentration range 102–
106 CFU/ml. The contents were mixed well and incubated for ∼20 h at 37 °C
and 1 ml aliquots were transferred into fresh 9 ml nutrient broth tubes, whereas
the remaining aliquots were pour plated with BHI agar and incubated for 24 h at
37 °C. The plates were observed for the bacterial colonies and the tubes for
turbidity [21]. Growth inhibitory activity was calculated using [(C − T) / C] × 100,
wherein C and T were the colony counts in the ‘control’ and ‘test’ plates,
respectively [5].
values calculated by viscometry. The DA calculated from IR
and 1H-NMR data was in good agreement with each other (Figs.
1 and 2, Table 1). The presence of a single broad peak centered
at 3371 cm− 1 in the IR-spectrum of native chitosan was indicative of its β-conformation [28]. The latter is more susceptible
for degradation due to parallel arrangement of the individual
chains, resulting in weak inter-chain hydrogen bonding network
and an easy penetration of the enzymes [3].
2.11. SEM
3.2. Pronase activity
After 20 h incubation at 37 °C, 0.5 ml aliquots from the above assay tubes
were transferred to micro-centrifuge tubes and centrifuged. The pellets obtained
were treated with phosphate buffer (pH 7.0, 0.3 M), fixed with glutaraldehyde
(1%) for 1 h at 4 °C and further treated with 10–96% alcohol in a sequential
manner and dried. They were spread on double-sided conducting adhesive tape
pasted on a metallic stub and coated with gold (100 μ) in a sputter coating unit
for 5 min and observed under SEM (LEO 435 VP, LEO Electron Microscopy
Ltd., Cambridge, UK) at 20 kV [22].
The purity of pronase was established unambiguously by
SDS-PAGE, capillary electrophoresis and also by reverse-phase
HPLC, wherein only a single peak was observed [14,29]. In
addition, N-terminal sequence analysis of pronase was found to
be Ser–Gln–Gly–Ser–Val–Tyr–XX–Pro–Tyr–Ala–Asp,
which was in harmony with the reported sequence data for
pronase. Optimum chitosan depolymerization by pronase
occurred at pH 3.5 and 37 °C and the reaction obeyed
Michaelis–Menten kinetics with Km and Vmax of 5.71 mg ml− 1
and 288.26 nmol min− 1 mg− 1, respectively [14]. The activity of
pronase towards chitosanolysis was 0.12 Unit (Unit = μmol of
reducing equivalents released min− 1 mg− 1) [30], whereas
towards proteolysis, it was 4 Units (Unit = enhanced absorbance
at 280 nm, due to the release of TCA soluble peptides from
hemoglobin ÷ reaction time × mg protein in the reaction mixture), indicating its preferential proteolysis. The enzyme was
stable up to 4 h at optimum conditions. Although pronase
showed increase in the activity, chitosan concentration above
20 mg ml− 1 (substrate: enzyme ratio of 200:1) could not be used
owing to its high viscosity, which affects enzyme penetration
and thus depolymerization.
2.12. Characterization of cell-free supernatant
(A) Cultures of B. cereus (106 CFU ml− 1) and E. coli (104 CFU ml− 1),
individually suspended in saline, were subjected to sonication (Julabo model
USR-1, West Germany; 51 kHz; 6 × 30 and 3 × 20 s, respectively) and incubated
at 37 °C for 24 h followed by centrifugation (10000 rpm, 4 °C for 15 min). The
supernatant was dialyzed (cut-off value 10 kDa) and freeze-dried (positive
control), (B) The cultures suspended in saline were centrifuged after incubation at
37 °C for 24 h, the supernatant was subjected to dialysis to remove saline and
freeze-dried (control) and (C) to one set of cultures in saline, was added LMWC
solution (1.0 and 3.0 mg ml− 1 of B. cereus and E. coli, respectively), incubated at
37 °C for 24 h and the supernatant obtained was dialyzed and freeze-dried (test).
The freeze-dried samples of A, B and C were analyzed for total carbohydrates and
proteins levels, and for cytoplasmic enzyme (maltase, lactase and protease)
activities [23–25], also subjected to native-PAGE [26]. Total carbohydrate is
measured by phenol-H2SO4 method (3 ml sample + 0.5 ml 5% phenol + 5 ml
concentrated H2SO4, after 15 min at room temperature, readings were taken as
490 nm) and the protein content was determined by using Bradford reagent (20 μl
sample + 1.8 ml Bradford reagent, readings were taken within 10 min at 595 nm).
Lactase was detected by the glucose oxidase-O-dianisidene method using
0.5 mM lactose in 0.2 M citrate phosphate buffer, pH 7.0, as substrate. Maltase
was detected by the glucose oxidase method using 5 mM maltose in 0.2 M citrate
phosphate buffer, pH 7.0, as substrate. In both, unit activity (U) is defined as the
amount of enzyme liberating 1 μg glucose in 1 min. Protease activity was
measured using casein as substrate. Enzyme solution (0.5 ml) suitably diluted
was mixed with 0.5 ml 100 mM Tris–HCl (pH 8.0) containing 1% casein, and
incubated for 15 min at 60 °C. The reaction was stopped by addition of 0.5 ml
TCA (20%, w/v). The mixture was allowed to stand at room temperature for
15 min and then centrifuged at 13,000 rpm for 15 min to remove the precipitate.
The absorbance was measured at 280 nm. A standard curve was generated using
solutions of 0–50 mg tyrosine. One unit of protease activity was defined as the
amount of enzyme, which liberated 1 μg tyrosine in 1 min at 60 °C. Native
(alkaline)-PAGE was performed using Tris–Glycine buffer of pH 8.3. The major
proteins on PAGE from positive control (A) and test (C) samples were blotted on
to PVDF membrane and subjected to N-terminal sequence analysis on an
Applied Biosystem 789 protein sequencer (USA). Fatty acid analysis was done
according to the method of Morrison and Smith [27].
3. Results and discussion
3.1. Characterization of native chitosan
The Mr of native chitosan was 71 ± 2 kDa as determined by
GPC on Sepharose CL-4B, which was in accordance with the
3.3. Characterization of LMWC
The Mr values calculated by HPLC (Table 1) and GPC were
in good agreement with each other, and varied between 9.5 and
8.5 kDa depending on the reaction time (i.e., 1–5 h). The use of
acetate buffer minimized interaction between the column
material and the free –NH2 groups on LMWC. Appearance of
a single symmetrical peak (Fig. 3) was indicative of molecular
homogeneity of the LMWC preparations.
3.3.1. IR spectroscopy
In the IR spectra of both chitosan and LMWC, the band
around 2900 cm− 1 due to the –CH vibration represented a good
internal reference for comparison with other band absorbance
(Fig. 1). The absence of a strong absorption band above
3500 cm− 1 indicated the involvement of both 3-OH and
primary hydroxyl group (6-CH2OH) in the intra- and inter-chain
hydrogen bonding [28]. A slight downward shift in the
absorption at around 3369 cm− 1 for LMWC (1 h) was
indicative of its disorderliness, whereas an upward shift
observed in 5 h sample indicated its orderliness. This could
be explained based on a fold-decrease in the DA although Mr of
1 and 5 h samples was comparable (Table 1). In case of 1 h
LMWC, fold-decrease in the Mr was ∼7.5, whereas that of DA,
498
A.B. Vishu Kumar et al. / Biochimica et Biophysica Acta 1770 (2007) 495–505
Fig. 1. IR spectra of native chitosan (A) and LMWC obtained after 1 h (B) and 5 h (C) of pronase catalyzed chitosanolysis.
∼ 1.35 compared to native chitosan, accounting to a decrease of
0.181 DA for every fold decrease in the Mr. Accordingly, if
pronase is specific for a type glycosidic linakge, then 5 h
LMWC with 8.35-fold decrease in the Mr was supposed to
show ∼1.51-fold decrease in the DA, but practically it showed
1.81-fold decrease, indicating that though pronase can act on all
the four types of glycosidic linkages in chitosan (–GlcN–
GlcN–, –GlcN–GlcNAc–, –GlcNAc–GlcN– as well as
–GlcNAc–GlcNAc–), it has more chitinase than chitosanaselike activity, releasing GlcNAc/GlcNAc-rich oligomers, leaving
more GlcN in the LMWC. The latter was also evident from our
earlier data [29], wherein there was the release of only GlcNAc
and not GlcN after chitosanolysis by pronase. To confirm
chitinase-like activity of pronase, we made use of chitosan
samples of Mr 71 ± 2 kDa having DA 15, 19, 26 and 40% with
which pronase showed activity of 0.71, 0.86, 1.12 and 1.52
Units, respectively.
Intensity of the peak near 1650 cm− 1 (amide I) in LMWC
decreased due to decrease in acetyl content. The absorbance
near 1320 cm − 1 , corresponding to N-acetylglucosamine
residues gradually decreased in LMWC (Absorbance = log %
T, 0.0668, 0.0290 and 0.0230, respectively for native chitosan,
LMWC-1 and 5 h) confirming a decrease in the DA. The band
around 1623 cm− 1 is expected to be associated with an intra-
A.B. Vishu Kumar et al. / Biochimica et Biophysica Acta 1770 (2007) 495–505
499
Fig. 3. HPLC profiles of LMWC obtained after 1 h, 3 h and 5 h of pronase
catalyzed chitosanolysis.
makes it difficult to assess the kind of chain packing or hydrogen bonding network [28].
Fig. 2. 1H-NMR spectra of native chitosan (A) and LMWC obtained after 1 h (B)
and 5 h (C) of pronase catalyzed chitosanolysis.
molecular hydrogen bond between –C = 0 and –NH2 groups. In
the present study, a decrease in the peak height near 1650 cm− 1
in LMWC (1 h) compared to native chitosan indicates a
decrease in the intra-molecular hydrogen bonding. In the 750–
500 cm− 1 region, thought to be more sensitive to changes in
crystallinity, a shoulder at 662.62 cm− 1 in case of 5 h LMWC is
assigned to out-of-plane bending of –NH and –OH groups and
participation in hydrogen bonding [28]. The region between
1460 and 1420 cm− 1 is considered for polysaccharides as
conformation sensitive and a shift of the same in LMWC is
attributed to modification in the environment of –CH2OH
group, although the high degree of band coupling in this region
3.3.2. Liquid state 1H-NMR
Lavertu et al. [18] performed spectral assay at 70 °C and found
that the hydrolytic cleavage of acetyl groups of chitosan by dilute
acid (i.e., DCl) at this temperature to be quite slow. However, in
the present study, as the native chitosan showed observable
cleavage of the acetyl group in the acidic medium above 60 °C,
the spectral measurements were performed at 23 °C, which also
showed the appearance of a solvent signal at around 4.37 ppm.
The DA calculated from 1H-NMR spectral data (Fig. 2) was
in accordance with that obtained from IR-spectra (Table 1). In
the spectra, the resonance near 1.8 ppm is assigned to the
N-acetyl protons and a progressive decrease in its height for 1 h
and 5 h LMWC indicated a decrease in DA. From the chemical
shift values (Table 2), it was obvious that depolymerization was
also associated with conformational changes, in accordance with
the IR spectra. Compared to native chitosan, 1 h samples showed
a downfield shift in the values, whereas 5 h LMWC showed an
upfield shift substantiating the orderliness and altered conformation of the latter, as also explained by IR-spectra.
3.4. Solubility of chitosan and LMWC
Solubility of chitosan plays a crucial role for its application,
which in turn depends on its Mr and DA. Chitosan is soluble in
Table 1
Characteristics and percent yield of LMWC
Chitosan
Native
LMWC—1 h
LMWC—3 h
LMWC—5 h
a
b
Mr (kDa)
71.0 ± 2.0a
9.50 ± 0.15b
9.10 ± 0.12b
8.50 ± 0.12b
DA (%)
IR
1
25.78
19.04
16.34
14.29
25.53
18.87
16.08
14.34
Determined by GPC on Sepharose CL-4B column.
Determined by HPLC.
H-NMR
Yield
(%)
–
80
77
74
Table 2
H-NMR—chemical shift values (in ppm) of native chitosan and LMWC
1
Chitosan
H-1(D)
H-2/6
H-3(D)
H-Ac
Native
LMWC—1 h
LMWC—5 h
4.65
4.58
4.73
3.53
3.51
3.61
2.92
2.91
2.93
1.84
1.82
1.89
500
A.B. Vishu Kumar et al. / Biochimica et Biophysica Acta 1770 (2007) 495–505
aqueous acidic medium below pH 6.2 and above this pH it
precipitates out of the solution. In the present study, native
chitosan was insoluble in aqueous medium (pH ∼ 5.6), showed
only 13% solubility in 0.01% acetic acid and its complete
dissolution required 1% acetic acid, while LMWC was 66–74 ±
2% soluble in aqueous medium and became completely
soluble in 0.01% acetic acid. To precipitate out LMWC after
pronase digestion, it was necessary to adjust the pH to 6.7 and
6.9 after 1 h and 5 h reaction time, respectively. Altogether,
solubility of chitosan/LMWC was not only dependent on pH of
the solvent used, but also the percent of acid. Surprisingly, the
5 h LMWC showed a decrease in solubility in aqueous medium
(66 ± 2%) compared to that obtained after 1 h (74 ± 2%
solubility). According to our earlier reports [9,29], observed
difference in the solubility of 1 h and 5 h samples could be due
to the removal of more of GlcNAc in 5 h sample, resulting in
LMWC having more GlcN, the –NH2 group of which undergo
stronger intra- and inter-chain hydrogen bonding, affecting the
solubility. Cheng and Li [9] made use of two volumes of
acetonitrile instead of equal volume of 2 N NaOH as in the
present study, to arrest the catalytic reaction. The use of
acetonitrile gives a product readily soluble in water, contrary to
the insoluble sodiated-form of LMWC. But the drawbacks of
using acetonitrile are its cost, flash treatment at elevated
temperature of the supernatant, which results in Maillard
reaction products and the consequent difficulty in recovering
both LMWC and chitooligomers-monomers in one step and the
safety of the resulting product due to the presence of residual
solvent.
Table 3
Inhibitory effect of chitosan and LMWC towards B. cereus and E. coli
Chitosan
Mr (kDa) DA Concentration Indicator bacteria
(%) (%)
B. cereus F4810a E. coli D21b
Percent inhibitionc
Native
71.0
26
LMWC-1 h
9.50
19
LMWC-3 h
9.10
16
LMWC-5 h
8.50
14
0.01
0.03
0.05
0.005
0.01
0.02
0.03
0.04
0.05
0.10
0.005
0.01
0.02
0.03
0.04
0.05
0.10
0.005
0.01
0.02
0.03
0.04
0.05
0.10
62 ± 2
61 ± 3
60 ± 2
94 ± 1
100 ± 0
100 ± 0
99 ± 1
98 ± 2
97 ± 2
98 ± 1
88 ± 2
98 ± 1
97 ± 1
96 ± 2
97 ± 1
97 ± 1
96 ± 2
86 ± 2
97 ± 1
95 ± 2
94 ± 2
95 ± 3
96 ± 2
94 ± 2
22 ± 1
20 ± 3
23 ± 2
20 ± 2
35 ± 3
58 ± 2
100 ± 0
98 ± 2
99 ± 1
97 ± 2
28 ± 1
46 ± 2
70 ± 1
100 ± 0
99 ± 1
96 ± 3
97 ± 1
40 ± 2
60 ± 1
75 ± 3
100 ± 0
98 ± 2
96 ± 2
97 ± 1
NI—no inhibition.
a
Inhibition was effective against a cell population of 106 CFU/tube.
b
Inhibition was effective against a cell population of 103 CFU/tube.
c
Values of mean of five trials ± standard deviation.
3.5. Bactericidal activity
Compared to native chitosan, LMWC showed better growth
inhibitory effect towards both Gram-positive and Gramnegative bacteria such as B. cereus, E. coli, Y. enterocolitica
and B. licheniformis. However, for further study, B. cereus and
E. coli, one each from Gram-positive and Gram-negative
bacteria, were chosen. The effects of Mr, DA and the
concentration of LMWC on growth inhibitory action are
given in Table 3. The growth inhibitory effect of LMWC was
more towards B. cereus as against E. coli. As could be seen
from Table 3, LMWC with Mr 9.5 kDa and 19% DA (1 h
sample) caused complete growth inhibition of B. cereus
whereas for E. coli, it was the 5 h sample (8.5 kDa and 14%
DA) that was more effective, indicating a differential role
played by both Mr and DA of LMWC. According to Chen et al.
[21], chitosans with lower DP and DA were more effective as
microbial growth inhibitory agents. The enhanced growth
inhibition and bactericidal activity associated with LMWC was
attributed to lower DA (∼ 14–19%), as a result of which, the
number of GlcN residues increases and at the assay condition
(pH 6.0), their amino group on C2 acquires a net positive charge
(–NH2 → –NH3+) and preferably bind to the negatively charged
microbial cell-surface causing its permeabilization. It has been
reported earlier, that chitosan of 3.1 kDa (DA, 52%) and
7.4 kDa (DA, 45%) showed only 58 and 71% inhibition,
respectively, towards B. cereus, confirming the dependency of
inhibition on the DA, Mr as well as the functional groups
present [2,4]. In the present study, the LMWC had a Mr ranging
between 9.5 and 8.5 kDa, which was comparable to those used
by the earlier workers, but even at 0.01% level, they showed
100% inhibition, which could be attributed to their relatively
low DA (∼ 14–19%), thus establishing its definite role for the
bactericidal activity [31].
From Table 3, it was clear that increase in the concentration
of LMWC showed a linear inhibitory effect on bacterial growth.
With B. cereus, LMWC obtained after 1 h reaction time showed
optimum lytic activity compared to 3 h and 5 h samples. Unlike
with B. cereus, 5 h LMWC showed better antibacterial activity
towards E. coli compared to 1 h and 3 h samples. This could be
due to the fact that 5 h sample (8.5 kDa, 13.4% DA), owing to
its smallest size and decreased DA can slide, so as to form
hydrogen bonding easily with the electronegative atoms of the
E. coli cell membrane that are present on the surface, unlike
with Gram-positive bacteria, wherein LMWC has to penetrate
the cell-wall in search of electronegative groups.
The MIC, defined as the lowest concentration of test samples
at which the cell growth is neither visible to naked eye
(turbidity) nor measurable by plating (viable counts), was found
to be 0.01% towards 106 CFU ml− 1 of B. cereus and 0.03%
towards 104 CFU ml− 1 of E. coli. Dilution of broth containing
the organism and test samples did not show any growth even
after prolonged incubation (for 96 h), indicating bactericidal
A.B. Vishu Kumar et al. / Biochimica et Biophysica Acta 1770 (2007) 495–505
effect of the LMWC, which was further confirmed by SEM
studies (Fig. 4), wherein disruption/disintegration of cell surface
could be seen.
Analyses of the cell-free supernatants of the sonicated and
LMWC treated cultures showed the presence of carbohydrates
and proteins (Table 4), in support of the observation of SEM.
The latter in the LMWC treated as well as sonicated culture
supernatants was several folds higher than that in the untreated
cultures. Similarly, increased levels of cytoplasmic enzymes
such as maltase, lactase and protease in the LMWC treated and
sonicated culture supernatants substantiated release of cytoplasmic contents (Table 4). Protein profiles (cell-free supernatants) on PAGE (Fig. 5) as well as HPLC (data not shown)
also supported the damaged cell wall/outer membrane,
evidenced by the release of proteins.
Gram-negative bacteria are known to have LPS and protein
in the outer membrane, whereas Gram-positive ones generally
have surface lipids [31]. Identification by GLC of fatty acids in
the LMWC-treated culture supernatants (Table 5) further
confirmed the dissociation and thus permeabilization of surface
lipids of B. cereus and the outer membrane of E. coli. The
501
Table 4
Constituents of cell free supernatants of controls and LMWC treated B. cereus
and E. coli
Samples
Carbohydrate
Protein
(μg/ml)
Maltase
Lactase
Protease
(Units)
B. cereus
Positive control (A)
Control (B)
Test (C)
49.8
0.74
44.1
124.8
23.58
109.8
0.16
0.03
0.14
0.28
0.02
0.26
0.30
0.01
0.28
E. coli
Positive control (A)
Control (B)
Test (C)
52.3
0.92
47.8
179.7
31.24
160.8
0.18
0.06
0.16
0.30
0.03
0.27
0.32
0.01
0.29
A—cultures subjected to sonication. B—cultures without any treatment. C—
cultures treated with LMWC.
release of saturated fatty acids in E. coli in abundance (14:0,
16:0 and 18:0 followed by 18, 20-series unsaturated fatty acids)
confirmed disruption of cell wall/membrane.
The amino acid sequence of the protein (electrophoretic
mobility, RF-0.15) from the supernatant of B. cereus treated
Fig. 4. SEM of B. cereus and E. coli—before (A and C) and after (B and D) treatment with LMWC, respectively (magnification, ×10000). E and F represent the SEM
immediately after the addition of LMWC to the test cultures (E—B. cereus and F—E. coli, magnification ×5000).
502
A.B. Vishu Kumar et al. / Biochimica et Biophysica Acta 1770 (2007) 495–505
Fig. 5. Native-PAGE of cell-free supernatants of B. cereus (A) and E. coli (B)
Lanes I and VI—cells subjected to sonication, II and V—untreated control, III
and IV—cells treated with LMWC. S—Protein standards.
with LMWC was found to be –Ala–Gly–Lys–Thr–Phe–Pro–
Asp–Val– and this was identified as a fragment of the
surface-layer (S-layer) protein of B. cereus (–X–Gly–Lys–
Thr–Phe–Pro–Asp–Val–). S-layer protein is a para-crystalline
mono-layered assembly of proteins, which coat the cell-surface
of B. cereus. Sequence of the protein from E. coli (RF-0.23)
determined by Edman degradation was –Ala–Pro–Val–Gly–
Ala– and this showed homology with a fragment of regulator of
nucleoside diphosphate kinase (rnk protein). Shankar et al. [32]
reported that the latter, a 14.0 kDa cytoplasmic protein with no
homology to any other protein, is involved in the restoration of
nucleoside diphosphate kinase activity. Identification of S-layer
and rnk proteins in the cell-free supernatants once again
confirmed pore formation and permeabilization of the bacterial
cell wall by the action of LMWC.
Further, the protein of RF 0.5 of both sonicated and LMWC
treated B. cereus culture showed identical N-terminal sequences
(–Ser–Lys–Phe–Gly–Leu–Pro–). BLAST analysis (www.
expasy.org) of this sequence indicated it to be a fragment of
nucleosomal enzyme, methionyl t-RNA formyltransferase [33].
The N-terminal sequence of the protein with RF of 0.42 from
sonicated E. coli culture was –Gln–Gly–Pro–Glu–Gln– and
that of LMWC treated was –Tyr–Asp–Pro–Glu–Ala–. Both
these proteins were fragments of the hypothetical cytoplasmic
protein ECs [34]. The appearance of different fragments of the
same protein could be attributed to the damage caused by
sonication and proteolysis.
3.6. Mechanism of bactericidal action of LMWC
Till date, the exact mechanism of growth inhibitory activity
of chitosans is yet to be elucidated. Stacking of chitosan
molecules over the microbial cell surface, thus blocking the
transport of nutrients [2] or binding to DNA and thus inhibiting
transcription or permeabilization of the microbial cell wall/
membrane or binding of trace metals or water binding and the
activation of several defense mechanisms in the host tissue, etc.
[3] have been postulated for antibacterial activity of chitosan.
Helander et al. [35] showed the binding of chitosan to the outer
membrane of Gram-negative bacteria forming a vesicular
structure, causing disruption and extensive alteration in the
outer membrane surface and resulting in the loss of its barrier
property. It is also possible that, due to its ability to bind LPS,
chitosan causes permeabilization of microbial cell membrane
and thus enhancing the non-protein nitrogenous substance
uptake and releasing the LPS from the cell surface [36]. We
could observe a sudden increase in the turbidity of the test
cultures immediately after the addition of LMWC and a steady
clearing of the solution, indicating initial interactions between
the two followed by cellular/membrane disintegration. The
former is also evident from the SEM of the cells immediately
after the addition of LMWC, wherein we could see irregularities
on the cell surface and aggregation of the cells (Fig. 4E and F).
Gram-negative bacteria contain an outer membrane wherein
LPS and proteins are held together by electrostatic interactions
with divalent metal ions, 1–2 layers of peptidoglycans (cellwall) and a cell membrane (containing lipid bilayer, transmembrane proteins and inner/outer membrane proteins).
Additionally there are present phospholipids, proteins and
lipoproteins. LPS are structurally characterized by three
independent regions namely, Lipid A, Core oligosaccharide
and O-specific chains. Lipid A contains repeating units of
(GlcN)2 to which fatty acids (mainly C10–C16) and also
phosphate groups are attached [37]. The Core region is
characterized by the presence of 2-keto-2-deoxyoctanoic acid
and heptoses. The O-specific antigenic oligosaccharide-repeating units, are made up of glucose, galactose including uronic
acid and some unusual sugars such as 2-O-acetyl abequose,
3-deoxyoctulosonic acid, 2-amino-2-deoxy-hexuronic acid,
4-O-lactylhexose, etc., which are negatively charged. These
negatively charged groups in E. coli cell wall/outer membrane
get involved in ionic-type of binding with –NH3+ groups of
LMWC (which are cationic in nature below pH 6.2) with
concomitant release of the LPS from the outer membrane,
distorting the electrostatic interactions of the latter with metal
ions, resulting in the exposure of cell-wall and cell membrane to
osmotic shock and thus spillage of the cytoplasm (Fig. 6A).
Earlier, we showed that the chitooligosaccharides (DP, 4–8) do
show bactericidal action towards Gram-negative bacteria owing
to their deposition on the bacterial cell-surface because of their
ionic interaction, thus blocking the nutrient flow [29]. Unlike
chitooligosaccharides, LMWC are strong enough to bind and
release the LPS from the Gram-negative bacterial cell wall,
which is also evident from the SEM (Fig. 4D) and this could be
due to the higher molecular structure of the LMWC.
In Gram-positive bacteria, the cell membrane is covered by a
cell-wall made up of 30–40 layers of peptidoglycans, which
Table 5
Fatty acid profiles of the cell-free supernatants of B. cereus and E. coli
Fatty acid
B. cereus
14:0
16:0
16:1
18:0
18:1
18:2
20:4
12.04
27.22
1.96
26.54
9.64
12.31
10.29
E. coli
Concentration (%)
1.90
40.34
33.32
6.46
15.33
1.34
1.31
A.B. Vishu Kumar et al. / Biochimica et Biophysica Acta 1770 (2007) 495–505
Fig. 6. Site of attachment of chitosan to outer membrane of Gram-negative bacterium (A) and cell wall of Gram-positive bacterium (B).
503
504
A.B. Vishu Kumar et al. / Biochimica et Biophysica Acta 1770 (2007) 495–505
contain repeating units of GlcNAc, N-acetylmuramic acid as well
as D and L-amino acids including isoglutamate and sometimes
toichoic acid, to which positively-charged –NH3+ groups of
LMWC could bind, resulting in cell-wall distortion–disruption
and exudation of the cytoplasmic contents (Fig. 6B). This was
evident from the SEM data (Fig. 4B), where one could see pores
along with irregularities on the cell surface of B. cereus and E. coli
after treatment with LMWC. The mode of action of cationic
antibacterial agents is widely believed to be due to interacting with
and disrupting the wall/membrane structure [29,38].
In conclusion, depolymerization of chitosan, associated with
decrease in Mr and DA, with the aid of pronase can be exploited
commercially for the bulk production of LMWC of desired
molecular weight. Pronase is easily available at a low cost
compared to chitosanase. LMWC of Mr, varying between 9.5
and 8.5 kDa and DA between 14 and 19% showed better
bactericidal activity towards both B. cereus and E. coli compared to chitosan of much higher Mr. Bactericidal activity was
essentially due to pore-formation on the cell surface and thus
permeabilization of the bacterial cell contents. Being non-toxic
LMWC could be of potential use as a food preservative.
Acknowledgements
Authors thank Ms. Padmaparna, Indian Institute of Science,
Bangalore, for 1H-NMR and Mr. Anbalagan, CIFS, CFTRI,
Mysore, for SEM. ABVK thanks CSIR, New Delhi, for Senior
Research Fellowship.
[11]
[12]
[13]
[14]
[15]
[16]
[17]
[18]
[19]
[20]
[21]
References
[22]
[1] J. Synowiecki, N.A. Khateeb, Production, properties and some new
applications of chitin and its derivatives, Crit. Rev. Food Sci. Nutr. 43
(2003) 145–171.
[2] D.H. Shon, Chitosan oligosaccharides for functional foods and microbial
enrichment of chitosan oligosaccharides in soy-paste, Proceedings of the
International workshop on Bioactive Natural Products, Tokyo, Japan,
2001, pp. 56–66.
[3] R.N. Tharanathan, F.S. Kittur, Chitin—The undisputed biomolecule of
great potential, Crit. Rev. Food Sci. Nutr. 43 (2003) 61–87.
[4] S. Sekiguchi, Y. Miura, H. Kancko, S.L. Nishimura, N. Nishi, M. Iwase, S.
Tokura, Molecular weight dependency of antimicrobial activity by
chitosan oligomers, in: E. Nishimuri, E. Doi (Eds.), Food Hydrocolloids:
Structure, Properties and Function, Plenum Press, New York, 1994,
pp. 71–76.
[5] Y.J. Jeon, P.J. Park, S.K. Kim, Antimicrobial effect of chitooligosaccharides produced by bioreactor, Carbohydr. Polym. 44 (2001) 71–76.
[6] Y. Kondo, A. Nakatani, K. Hayashi, M. Ito, Low molecular weight
chitosan prevents the progression of low dose streptozotocin induced
slowly progressive diabetes mellitus in mice, Biol. Pharmacol. Bull. 23
(2000) 1458–1464.
[7] G.J. Tsai, Z.Y. Wu, W.H. Su, Antibacterial activity of a chitooligosaccharide mixture prepared by cellulase digestion of shrimp chitosan and its
application in milk preservation, J. Food Prot. 63 (2000) 747–752.
[8] S.C.W. Richardson, H.V.J. Kolbe, R. Duncan, Potentials of low molecular
mass chitosan as a DNA delivery system: biocompatibility, body distribution and ability to complex and protect DNA, Int. J. Pharm. 178 (1999)
231–243.
[9] C.Y. Cheng, Y.K. Li, An Aspergillus chitinase with potential for largescale preparation of chitosan oligosaccharides, Biotechnol. Appl. Biochem. 32 (2000) 197–203.
[10] K. Ogawa, O.A. Chrispinas, N. Yoshida, J. Inoue, K. Kariya, Chitosanase,
[23]
[24]
[25]
[26]
[27]
[28]
[29]
[30]
[31]
[32]
its manufacture, and manufacture of chito-oligosaccharides, Jpn. Kokai
Tokyo Koho, JP (2001) 0069975.
E.S. Santos, R. Guirardello, T.T. Franco, Purification of chitosanase from
Bacillus cereus, in: R.A.A. Muzzarelli (Ed.), Chitin Enzymology, ATEC,
Italy, 2001.
F.S. Kittur, A.B. Vishu Kumar, L.R. Gowda, R.N. Tharanathan,
Chitosanolysis by a pectinase isozyme of Aspergillus niger—A nonspecific activity, Carbohydr. Polym. 53 (2003) 191–196.
A.B. Vishu Kumar, M.C. Varadaraj, R.G. Lalitha, R.N. Tharanathan, Low
molecular weight chitosan: preparation with the aid of papain and
characterization, Biochim. Biophys. Acta 1670 (2004) 137–146.
A.B. Vishu Kumar, L.R. Gowda, R.N. Tharanathan, Non-specific
depolymerization of chitosan by pronase and characterization of the
resultant products, Eur. J. Biochem. 271 (2004) 713–723.
F.S. Kittur, K.R. Kumar, R.N. Tharanathan, Functional packaging
properties of chitosan films, Z. Lebensm. Unter. Forsch., A 206 (1998)
44–47.
W. Wang, B. Shuqin, L. Shuqing, Q. Wen, Determination of the Mark–
Houwink equation for chitosans with different degrees of deacetylation,
Int. J. Biol. Macromol. 13 (1991) 281–285.
D. Baxter, Michael, K.D.A. Taylor, Improved method for i.r. determination
of the degree of N-acetylation of chitosan, Int. J. Biol. Macromol. 14
(1992) 66–69.
M. Lavertu, Z. Xia, A.N. Serrequ, M. Berrada, A. Rodrigues, D. Wang, M.
D. Buschmann, A. Gupta, A validated 1H-NMR method for the determination of the degree of deacetylation of chitosan, J. Pharm. Biomed.
Anal. 32 (2003) 1149–1158.
C. Qin, D. Yumin, Z. Lintao, Z. Fangang, L. Yi, B. Zhou, Effect of
hemicellulase on the molecular weight and structure of chitosan, Polym.
Degrad. Stab. 80 (2003) 435–441.
S. Bharathi, M.N. Ramesh, M.C. Varadaraj, Predicting the behavioral
pattern of Escherichia coli in minimally processed vegetables, Food
Control 12 (2001) 275–284.
C.S. Chen, W.Y. Lian, G.J. Tsai, Antimicrobial effects of N-sulfonated and
N-sulfobenzyl chitosan and application to oyster preservation, J. Food
Prot. 64 (1998) 1124–1128.
L.A. Mc Dougall, W.H. Holzapfer, U. Schilinger, D.E. Felly, J.H. Rupnow,
Scanning electron microscopy of target cells and molecular weight
determination of a bacteriocin produced by Lactococcus lactis D53, Int. J.
Food Microbiol. 24 (1994) 295–308.
O. Olusanya, Detection of carbohydrases as a possible method of
differentiating faecal and non-faecal E. coli, FEMS Microbiol. Lett. 30
(1985) 277–281.
B. Ghorbel, B. Sellami-Kamoun, M. Nasri, Stability studies of protease
from Bacillus cereus BG1, Enzyme Microb. Technol. 32 (2003)
513–518.
F. Maccari, N. Volpi, Detection of submicrogram quantities of Escherichia
coli lipopolysaccharides by agarose-gel electrophoresis, Anal. Biochem.
322 (2003) 185–189.
U.K. Laemmli, Cleavage of structural proteins during the assembly of head
of Bacteriophage T4, Nature 227 (1970) 680–685.
M.R. Morrison, M. Smith, Preparation of fatty acid methyl esters and
dimethyl acetyls from lipids with boron fluoride methanol, J. Lipid Res. 5
(1963) 600–608.
B. Focher, A. Naggi, G. Torri, A. Cosani, M. Terbojevich, Chitosans from
Euphausia superba. 2: Characterization of solid-state structure, Carbohydr.
Polym. 18 (1992) 43–49.
A.B. Vishu Kumar, M.C. Varadaraj, R.G. Lalitha, R.N. Tharanathan,
Characterization of chitooligosaccharides prepared by chitosanolysis
with the aid of papain and pronase, and their bactericidal action
against Bacillus cereus and Excherichia coli, Biochem. J. 391 (2005)
167–175.
T. Imoto, K. Yagishita, A simple activity measurement of lysozyme, Agric.
Biol. Chem. 33 (1971) 1154–1157.
H.K. No, N.Y. Park, S.H. Lee, S.P. Meyers, Antimicrobial activity of
chitosans and chitosan oligomers with different molecular weights, Int. J.
Food Microbiol. 74 (2002) 65–72.
S. Shankar, D. Schlictman, A.M. Chakrabarty, Regulation of nucleoside
A.B. Vishu Kumar et al. / Biochimica et Biophysica Acta 1770 (2007) 495–505
diphosphate kinase and an alternative kinase in Escherichia coli: role of
the sspA and rnk genes in nucleoside triphosphate formation, Mol.
Microbiol. 17 (1995) 135–943.
[33] N. Ivanova, A. Sorokin, I. Anderson, N. Galleron, B. Candelon, V.
Kapatral, A. Bhattacharya, G. Reznik, N. Mikhallova, A. Lapidus, L. Chu,
M. Mazur, E. Goltsman, N. Larsen, M. C'souza, T. Walunas, Y. Grechkin,
G. Pusch, R. Hoselkorn, M. Fonstein, S.D. Ehrlich, R. Overbeek, N.
Kyrpides, Genome sequence of Bacillus cereus and comparative analysis
with Bacillus anthracis, Nature 423 (2003) 87–91.
[34] R.A. Welch, V. Burland, G. Punkett, P. Redford, P. Roesch, D. Rasko, E.L.
Buckles, S.R. Liou, A. Boutin, J. Hackett, D. Stroud, G.F. Mayhew, D.J.
Rose, S. Zhou, D.C. Schwartz, N.T. Perna, H.L.T. Mobley, M.S.
Donnenberg, F.R. Blattner, Extensive mosaic structure revealed by the
[35]
[36]
[37]
[38]
505
complete genome sequence of uropathogenic Escherichia coli, Proc. Natl.
Acad. Sci. NY 99 (2002) 17020–17024.
I.M. Helander, E.L. Nurmiaho-Lassila, R. Ahvenainen, J. Rhoades, S.
Roller, Chitosan disrupts the barrier properties of the outer membrane of
Gram-negative bacteria, Int. J. Food Microbiol. 71 (2001) 235–244.
X.F. Liu, Y.L. Guan, D.Z. Yang, Z. Li, K. De Yao, Antibacterial action of
chitosan and carboxymethylated chitosan, J. Appl. Polym. Sci. 79 (2001)
1324–1335.
H. Mayer, R.N. Tharanathan, J. Weckesser, Analysis of lipopolysaccharides
of Gram-negative bacteria, Methods Microbiol. 18 (1985) 157–207.
Y. Nakajima, J. Ishibashi, F. Yukuhiro, A. Asaoka, D. Taylor, M. Yamakawa,
Antibacterial activity and mechanism of action of tick defensin against
Gram-positive bacteria, Biochim. Biophys. Acta 1624 (2003) 125–130.