PHYTOTHERAPY RESEARCH

Phytother. Res. 27: 1517–1523 (2013)

Published online 28 November 2012 in Wiley Online Library
(wileyonlinelibrary.com) DOI: 10.1002/ptr.4891

Effect of Corilagin on Membrane Permeability of

Escherichia coli, Staphylococcus aureus and
Candida albicans

Na Li,1 Meng Luo,1 Yu-jie Fu,1* Yuan-gang Zu,1 Wei Wang,1 Lin Zhang,1 Li-ping Yao,1
Chun-jian Zhao1 and Yu Sun2
1
Engineering Research Center of Forest Bio-Preparation, Ministry of Education, Northeast Forestry University, Harbin 150040, PR
China
2
Department of General Surgery, The Third Affiliated Hospital of Liaoning Medical University, Jinzhou 121000, PR China

Corilagin is a member of polyphenolic tannins. Its antimicrobial activity and action mechanism against Escherichia
coli, Staphylococcus aureus and Candida albicans were investigated through membrane permeability. Crystal
violet staining determination, outer membrane (OM) and inner membrane (IM) permeability, sodium dodecyl
sulfate polyacrylamide gel electrophoresis (SDS-PAGE) and atomic force microscopy (AFM) were used as
methods for our investigation. The minimum inhibitory concentrations were 62.5, 31.25 and 62.5 mg/mL for
E. coli, S. aureus and C. albicans, respectively. Crystal violet results and SDS-PAGE of supernatant proteins
showed that corilagin dose-dependently affected membrane permeability of E. coli and C. albicans but not of
S. aureus. OM and IM permeability assays revealed comparable results for E. coli. By using AFM, we demon-
strated extensive cell surface alterations of corilagin-treated E. coli and C. albicans. SDS-PAGE of precipi-
tated proteins revealed possible targets of corilagin, i.e. Fib, Sae R, Sar S in S. aureus and Tye 7p in C.
albicans. In conclusion, corilagin inhibited the growth of E. coli and C. albicans by disrupting their membrane
permeability and that of S. aureus by acting on Fib, Sae R and Sar S but not on membrane integrity. Copyright
© 2012 John Wiley & Sons, Ltd.
Keywords: corilagin; antimicrobial activity; mechanism; membrane permeability; protein.

Abbreviations: MIC, minimum inhibitory concentration; ONPG, ο-nitrophenyl-b-D-galactoside; PADAC, 7-[(thienyl-2-acetamido)-3-

[2-(4-N, N-dimethylamino phenylazo)-pyrididium-methyl)-3-cephem-4-carboxylic acid]; OM, outer membrane; IM, inner membrane;
AFM, atomic force microscopy.

or inhibiting microbial metabolism through oxidative

INTRODUCTION
E. coli, S. aureus and C. albicans represent exemplary
model organisms for gram-negative bacteria, gram-
positive bacteria and fungi which can induce numerous
diseases in humans. Broad-spectrum antimicrobial
agents can eradicate these diseases, but can also result
in drug resistance and severe side effects (Prasad et al.,
1995). Natural products which avoid drug resistance
and reveal low toxicity and high efficiency are in the
center of interest for treating infectious diseases. A ma-
jority of antimicrobial phytochemicals are derived from
the chemical classes of phenolics, terpenoids, alkaloids,
lectins, polypetides or polyacetylenes (Cowan, 1999).
Tannins are polyphenols, which can inhibit the growth
of microorganisms. In many fruits, they are regarded
as natural defense mechanisms against microbial infec-
tions (Chung et al., 1998). Tannins exert antimicrobial
activity by binding to cell walls, disrupting membrane
integrity, binding to proteins, inhibiting enzyme activity
* Correspondence to: Yujie Fu, Ph. D, Professor, Vice Director. Key
Laboratory of Forest Plant Ecology, Ministry of Education, Northeast
Forestry University, Box 332, Hexing Road 26, Harbin 150040, PR China.
E-mail: yujie_fu2002@yahoo.com
Copyright © 2012 John Wiley & Sons, Ltd.
phosphorylation (Cowan, 1999).
Corilagin (beta-1-O-galloyl-3, 6-(R)-hexahydroxydi-
phenoyl-D-glucose) is a polyphenolic tannin. This
compound possesses considerable hepatoprotective, anti-
HIV, thrombolytic, antihypertensive, anti-inflammatory,
antineoplastic, anti-oxidative and antimicrobial activities
(Zhao et al., 2008). In the past years, the antimicrobial
activity of corilagin has been intensively investigated.
The inhibition zone diameters of corilagin to S. aureus
and C. albicans were 12 mm for 100 mg, which were larger
than those of mallotinic acid, mallorepanin, chebulagic
acid, ellagic acid and gallic acid (Fogliani et al., 2005),
indicating a superior antimicrobial activity of corilagin.
However, the mode of action of corilagin towards micro-
organisms has not been elucidated yet. Therefore, E. coli,
S. aureus and C. albicans were chosen in the present study
to investigate the role of membrane permeability for the
mechanism of corilagin.
Corilagin was investigated by twofold serial broth dilu-
tions. The crystal violet assay, outer membrane (OM)
and inner membrane (IM) permeability assays were
carried out to analyze changes of membrane permeability.
It is widely accepted that the release of intracellular pro-
teins is another hallmark for membrane damage and the
loss of membrane integrity (Vaara and Vaara, 1981). Thus,
sodium dodecyl sulfate polyacrylamide gel electrophoresis
Received 9 May 2012
Revised 17 October 2012
Accepted 24 October 2012
1518 N. LI ET AL.
(SDS-PAGE) of supernatant proteins was conducted to
further demonstrate treatment-related alteration. If cell
membranes are disrupted and the membrane permeability
is changed, more proteins will be released into the super-
natant of corilagin-treated cell cultures. Atomic force
microscopy (AFM) is a suitable technique for determining
cellular morphology. Here, it was used to investigate
changes of membrane permeability and morphology
(Devi et al., 2010). The morphological parameters
including length, width, height and the calculated root-
mean-square (RMS) were also obtained from AFM.
These parameters provided information on how corilagin
treatment affected morphology and structure of microbial
cells. To study whether corilagin acts on certain target
proteins of microorganisms, SDS-PAGE of whole-cell
protein lysates was performed by using a recently
published assay (Zu et al., 2010). Then, alterations of
protein bands between corilagin-treated and untreated
cells were compared. However, the exact protein action
targets could not be found without introducing further
methods, such as two-dimensional electrophoresis. From
this point of view, more investigation is required.
MATERIALS AND METHODS
Materials. Corilagin (purity > 99%) was provided by
Shanghai Winherb Medical Science Co., Ltd. Erythro-
mycin, streptomycin and nystatin (all purity > 99%) were
used as positive control and were purchased from the
Sigma Chemical Co. (St. Louis, Mo, USA). Crystal
violet, ο-nitrophenyl-b-D-galactoside (ONPG), 7-[(thienyl-
2-acetamido)-3-[2-(4-N, N-dimethylamino phenylazo)-
pyrididium-methyl)-3-cephem-4-carboxylic acid] (PADAC)
were obtained from Sigma Chemicals (Shanghai, China).
30% Acr-Bis (29:1), Tris-HCl, pH 8.8, Tris-HCl, pH 6.8,
10% SDS, ammonium persulfate and TEMED were
purchased from Sigma Chemicals (Beijing, China) and
stored at 4
C except of ammonium persulfate, which was
confected to 10% and stored at -20
C.
Microorganisms and culture conditions. Escherichia coli
(ATCC 8739), Staphylococcus aureus (ATCC 6538) and
Candida albicans (ATCC 10231) were purchased from
the Institute of Applied Microbiology, Heilongjiang
Academy of Science (China). They were kept on agar
slants at 4
C and cultured on nutrient agar at 37
C or
28
C for 24 h. Prior to each experiment, E. coli and
S. aureus were refreshed from the nutrient agar in nutri-
ent broth at 37
C, and C. albicans in YPD broth at
28
C, and grown to a mid-log phase for further study
(Demirci et al., 2008).
Evaluation of antimicrobial activity. Microbial cells
(105 CFU/mL) were inoculated into a nutrient broth or
YPD broth at 0.1 mL/well, in 96-well microtiter plates.
Minimum inhibitory concentrations (MICs) were tested
by means of serial twofold dilutions following methods
recommended by the Clinical and Laboratory Standards
Institute. The stock solution of corilagin was 500 mg/mL
in 0.5% dimethyl sulfoxide (DMSO), and the required
concentrations for experiments were prepared from this
stock solution. Twofold serial dilutions of corilagin using
sterile distilled water were prepared from 500 mg/mL to
1.95 mg/mL in test tubes and then transferred into
Copyright © 2012 John Wiley & Sons, Ltd.
96-well microtiter plates. Microbial suspensions were
standardized to 108 CFU/mL. Freshly prepared micro-
bial suspensions were pipetted into each well in an equal
volume. After 24 h of incubation at 37
C for E. coli and
S. aureus, and 28
C for C. albicans, MICs were deter-
mined by visually inspecting each well for microbial
growth. The lowest corilagin concentration that resulted
in no microbial growth was taken as MIC. Erythro-
mycin, streptomycin and nystatin were used as positive
controls. All determinations were performed in
duplicate.
Crystal violet assay. The variation of E. coli, S. aureus and
C. albicans in membrane permeability was investigated by
crystal violet assay (Vaara and Vaara, 1981). Microbial
cultures grown to 108 CFU/mL were harvested by centri-
fugation at 4500  g, 4
C for 5 min twice. The precipita-
tions were resuspended in sterile 0.5% NaCl solution
and mixed with corilagin (500 mg/mL in 0.5% DMSO) to
a final corilagin concentration of 15.625 mg/mL and then
incubated at 37
C or 28
C for 8 h. Cells without corilagin
treatment were regarded as blank control. Erythromycin
and streptomycin were selected as positive control for
E. coli and S. aureus, and nystatin for C. albicans. Then,
cells were harvested at 9300  g for 5 min and
resuspended in 0.5% NaCl solution containing 10 mg/mL
crystal violet. They were incubated at 37
C for 10 min,
shaked and centrifuged at 13400  g for 15 min. The
supernatant were measured by UV–VIS spectrophotom-
eter at OD590. The OD values of the crystal violet-stained
initial solution used in the assay were investigated and
considered as 100%. All determinations were done in
triplicates. The percentages of crystal violet uptake were
calculated using following formula: (OD value of the
sample)/(OD value of crystal violet solution)  100
OM and IM permeability assays. OM and IM permeabil-
ities of E. coli were determined by measuring the release
of b-lactamase and cytoplasmic b-galactosidase activity,
respectively, from E. coli into the culture medium using
PADAC (7-[(thienyl-2-acetamido)-3-[2-(4-N, N-dimethy-
lamino phenylazo)-pyrididium-methyl)-3-cephem-4-
carboxylic acid]) or ONPG (ο-nitrophenyl-b-D-galactoside)
as substrate (Lehrer et al., 1988). The bacteria were grown
in nutrient broth which contained 2% lactose for IM
permeability assay and adjusted to 108 CFU/mL. Cells were
harvested by centrifugation at 9300  g for 10 min, washed
and then resuspended in 0.5% NaCl solution. Then, 100
mL corilagin (62.5 mg/mL and 250 mg/mL) was added to a
well containing 100 mL bacterial suspension, followed by
12 mL PADAC (50 mM) or ONPG (2.5 mM) added,
respectively, to each well. Control samples were similarly
prepared without corilagin treatment. Streptomycin at
50 mg/mL was used as positive control. The hydrolysis of
PADAC or ONPG over time was determined by measuring
the increase in A420 using a UV–VIS spectrophotometer. All
tests were done in triplicate.
SDS-PAGE of supernatant proteins and whole-cell
protein lysates. Logarithmically growing cells of E. coli,
S. aureus and C. albicans (each adjusted to 108 CFU/mL)
were treated with 500 mL corilagin at MIC and 4xMIC
for 8 h at 37
C or 28
C, respectively. Controls were run
without corilagin treatment. The cells were harvested by
centrifugation at 13400  g for 10 min at 4
C. Precipita-
tions of whole-cell protein lysates were washed twice with
Phytother. Res. 27: 1517–1523 (2013)
 ANTIMICROBIAL ACTIVITY OF CORILAGIN
phosphate buffer saline (PBS 50 mM; pH 7.0). The protein
fraction contained in the supernatant was condensed using
the TCA/acetone. Subsequently, 100 mL and 20 mL of the
sample buffer (0.06 M Tris-HCl (pH 6.8), 5% mercap-
toethanol, 10% glycerol, 2% SDS, 0.001% bromophenol
blue) were added to the precipitated and concentrated
proteins, respectively, and boiled for 10 min. Besides, the
concentrations of whole-cell protein lysates and super-
natant proteins in the samples were estimated by Lowry’s
method and adjusted to the same concentration for
all samples of each microorganism. The SDS-PAGE was
performed with a 5% stacking gel and a 12% separating
gel followed by silver staining (Laemmli, 1970; Spanos
et al., 1981).
Morphology changes of microorganisms. The mor-
phology change of E. coli, S. aureus and C. albicans cells
induced by corilagin was examined by AFM (Braga and
Ricci, 1998). Microorganisms treated with corilagin with
MIC and 4xMIC were incubated for 8 h at 37
C or
28
C. Then, cells were harvested by centrifugation at
9300  g for 10 min at 4
C, washed twice with PBS
(50 mM; pH 7.0) and subsequently resuspended in
PBS. The untreated cells were regarded as control. For
AFM analysis, 10 mL of the samples was applied on a
freshly cleaved mica surface and dried for 5 min before
imaging. Then, the mica surface containing samples
was softly washed with deionized water for three times
and dried in the air for examining. Individual microor-
ganisms of each sample solution were randomly selected
and imaged by AFM. All images shown in the present
study were representative. Furthermore, for each
group of cells, the morphological parameters including
length, width and height were precisely measured using
the off-line analysis commands provided by NanoScope
Software. To quantitatively describe the topography of
the bacterial surface, average surface roughness was
calculated using RMS average of the heightffi deviations.
vffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffiffi
u N
uX ðZi  ZmÞ2
The equation was Rrms ¼ t , where N is
i¼1
ðN  1 Þ
the total number of data points, Zi is the height of the
ith point and Zm is the mean height (Girasole et al.,
2007). A fixed size of 500 nm2 of microbial surface was
measured for the determination of roughness values.
At least 30 cells were processed to calculate mean
values for the each parameter.
RESULTS AND DISCUSSION
The antimicrobial activity of corilagin was investigated
against E. coli, S. aureus and C. albicans. As shown in
Table 1, MICs of corilagin were 62.5 mg/mL, 31.25 mg/mL
and 62.5 mg/mL for E. coli, S. aureus and C. albicans,
respectively (Table 1). In comparison with erythromycin
as positive control, stronger antibacterial activity to E. coli
was observed with corilagin. Meanwhile, corilagin
had higher antimicrobial activity than streptomycin
towards S. aureus and C. albicans. Corilagin was also
more effective than some other compounds with similar
structure, such as mallotinic acid, mallorepanin, chebu-
lagic acid, ellagic acid or gallic acid (Fogliani et al.,
2005). Corilagin consists of phenolic groups and
Copyright © 2012 John Wiley & Sons, Ltd.
1519
Table 1. MICs (mg/mL) of corilagin against microorganism.
MICs (mg/mL)
Microorganism Corilagin Erythromycin Streptomycin Nystatin
E. coli 62.5 125 31.25 —
S. aureus 31.25 1.95 62.5 —
C. albicans 62.5 7.82 250 20
aromatic rings and possesses astringency. It has been
proved that the astringent character of tannins is
responsible for complexation with enzymes leading to
inhibition of growth of microorganisms (Scalbert, 1991).
Therefore, the antimicrobial activity of corilagin may be
due to its astringent features. However, the action
mechanism has not been thoroughly investigated and
the role of membrane permeability for corilagin’s activity
is unknown.
As a structural component, the cytoplasmic cell mem-
brane is target for many antimicrobial agents. Crystal
violet poorly penetrates the membrane, but it easily
enters, if the membrane is defective (Devi et al., 2010).
As shown in Fig. 1, crystal violet uptake of E. coli, S. aureus
and C. albicans was 31.16%, 19.9% and 31.16%, respec-
tively, in the absence of corilagin and increased to 77.94%,
21.91% and 57.92%, respectively, after corilagin treatment
at MIC. At 500 mg/mL, it increased to 85.06%, 30.61% and
75.94%, respectively. A significant enhancement of crystal
violet uptake was observed in E. coli and C. albicans but
not in S. aureus. These effects on membrane permeability
were probably due to differences on membrane structure
and composition of the three microorganisms (Xing et al.,
2009; Cho and Lee, 2011), indicating different modes of
action of corilagin.
One of the structural discrepancies between Gram-
positive and Gram-negative bacteria is the composition
of the membrane. Gram-positive bacteria possess one
lipid layer, whereas Gram-negative bacteria possess two
cell membranes, namely OM and IM. b-lactamase, a peri-
plasmic enzyme and b-galactosidase, a cytoplasmic en-
zyme are both present in Gram-negative bacteria
(Nagaoka et al., 2000). b-lactamase which is located in
the periplasmic space is cryptic for PADAC, unless the
OM is permeabilized. Similarly, b-galactosidase which is
in cytoplasm cannot hydrolyze ONPG, until its inner
membrane undergoes permeabilization (Thennarasu
and Nagaraj, 1996). Consequently, PADAC and ONPG
which are impermeable to outer and inner membranes
were used as substrates to investigate the changes of
OM and IM permeability of E. coli (Thennarasu and
Nagaraj, 1996). If OM and IM were damaged and
functionally inactive, b-lactamase and b-galactosidase
activity could be released out of the cells or permeated
the cytoplasmic membrane, respectively. The ability of
b-lactamase and b-galactosidase activity to effuse OM
and IM of E. coli was evaluated by the production of
PADAC and ONPG hydrolysis followed by UV spectros-
copy (Ibrahim et al., 2001). As shown in Fig. 2, corilagin
caused a progressive concentration-dependent release of
b-lactamase and b-galactosidase activity into the medium
within 6 h, which was in accordance with the results of the
crystal violet assay. Corilagin may bind to OM and IM of
E. coli, disrupt membrane integrity and result in the loss
of barrier function or prevent the nutrient flow with
concomitant bacterial inhibition.
Phytother. Res. 27: 1517–1523 (2013)
1520 N. LI ET AL.

Figure 1. Crystal violet uptake of E. coli, S. aureus and C. albicans after corilagin treatment (diluted from 500 mg/mL to 15.625 mg/mL in two
folds) for 8 h. EM (Erythromycin) and SM (Streptomycin) were selected as positive control to E. coli and S. aureus, and NS (Nystatin) to C.
albicans.

Moreover, the release of intracellular proteins is
another hallmark for membrane damage and loss of
membrane integrity (Vaara and Vaara, 1981). Thus,
SDS-PAGE of supernatant proteins was carried out for
further verification of the changes of membrane perme-
ability. Figure 3 exhibits the invariability of supernatant
proteins of S. aureus (Fig. 3-C) and increment of E. coli
and C. albicans in dose-dependent manner (Fig. 3-A, E).
It proved the alteration of membrane permeability in E.
coli and C. albicans but not in S. aureus, which was in
accordance with the results of the crystal violet assay
and the determination of OM and IM permeability.
SDS-PAGE of whole-cell protein lysates was compared
with the National Center for Biotechnology Information

Figure 2. The changes of outer (A) and inner (B) membrane permeability of E. coli treated by various concentrations of corilagin (0, 62.5,
250 mg/mL) for 8 h. Streptomycin (50 mg/mL) was selected as positive control. PADAC and ONPG were used as substrate, respectively. The
release of periplasmic b-lactamase and cytoplasmic b-galactosidase activity was measured by using UV–VIS spectrophotometer at A420 nm.

Copyright © 2012 John Wiley & Sons, Ltd. Phytother. Res. 27: 1517–1523 (2013)
 ANTIMICROBIAL ACTIVITY OF CORILAGIN
Figure 3. SDS-PAGE of supernatant and precipitated proteins from
E. coli (A, B), S. aureus (C, D) and C. albicans (E, F) during corilagin treat-
ment. A: Lane-1: untreated control; Lane-2: 62.5 mg/mL corilagin;
Lane-3: 250 mg/mL corilagin; Lane-4: molecular marker. B: Lane-1: mo-
lecular marker; Lane-2: untreated control; Lane-3: 62.5 mg/mL corilagin;
Lane-4: 250 mg/mL corilagin. C: Lane-1: 31.25 mg/mL corilagin; Lane-2:
125 mg/mL corilagin; Lane-3: untreated control; Lane-4: molecular mar-
ker. D: Lane-1: molecular marker; Lane-2: untreated control; Lane-3:
125 mg/mL corilagin; Lane-4: 31.25mg/mL corilagin. E: Lane-1: 250mg/mL
corilagin; Lane-2: 62.5 mg/mL corilagin; Lane-3: untreated control;
Lane-4: molecular marker. F: Lane-1: molecular marker; Lane-2: untreated
control; Lane-3: 62.5 mg/mL corilagin; Lane-4: 250 mg/mL corilagin. This
figure is available in colour online at wileyonlinelibrary.com/journal/ptr.
Copyright © 2012 John Wiley & Sons, Ltd.
1521
to identify possible target proteins of corilagin. The
expression levels of Fib (19 kDa), Sae R (26 kDa), Sar S
(29.9 kDa) of S. aureus (Fig. 3-D) and Tye 7p (30.5 kDa)
of C. albicans (Fig. 3-F) decreased in corilagin-treated
cells compared with the untreated control. Fib (fibrino-
gen-binding protein) is unique to S. aureus and one of its
virulence factors (Boden Wastfelt and Flock, 1995). Sae
R belongs to the Sae family, whose proteins are important
for mediating a variety of environmental conditions,
including the resistance against antibiotics, high concen-
trations of NaCl and low pH (Rajan and Richard, 2008).
Having in mind that Sae R is a member of the Sae family,
the decrease of SaeR after corilagin treatment might be
correlated with its inhibition of S. aureus growth. In
addition, Sar S can induce the expression of protein A,
which is one of virulence determinants in S. aureus (Li
et al., 2003). Thus, a decrease of Sar S expression may
suppress the activity of protein A and, hence, the viru-
lence of S. aureus to humans or animals. In summary,
the decrease of Fib, Sae R and Sar S expression may
reduce the adaptive capacity of S. aureus to different
environmental conditions. Nevertheless, expression of
these proteins kept constant in E. coli after corilagin treat-
ment (Fig. 3-B). In C. albicans, Tye7p and Gal4p are two
key players of glycolytic transcriptional regulation and
are important for the fermentative growth of C. albicans
(Askew et al., 2009). Therefore, the decrease of Tye 7p ex-
pression may inhibit the energy metabolism of C. albicans
leading to growth inhibition. In conclusion, the growth of
S. aureus and C. albicans may be inhibited by corilagin
via either affecting protein synthesis or gene expression. In-
hibition might also be attributed to non-specific forces such
as hydrogen bonding and hydrophobic effects, as well as
covalent bond formation to complex with proteins as
described for tannins (Haslam, 1996). However, more
investigations are needed to verify this speculation.
Nowadays, AFM has become a valuable tool in
microbiological research. To further analyze the
changes of membrane permeability, AFM was used to
study the influence of corilagin on E. coli, S. aureus
and C. albicans under nearly physiological conditions.
Untreated cells of E. coli, S. aureus and C. albicans
(Fig. 4, control-A, B, C) had reasonably structured
surfaces without notable ruptures or large pores. How-
ever, corilagin caused irregularity on the surfaces of E.
coli and C. albicans at MIC (Fig. 4, MIC-A, C) and even
lost their membrane integrity (Fig. 4, 4MIC-A, C). The
treated microorganisms which lost their native shape
revealed desquamated cell walls, effusive cytoplasms
and finally underwent microbial death (Fu et al., 2007).
Thus, the loss of membrane integrity is responsible for
growth inhibition of E. coli and C. albicans by changing
membrane permeability. Similar results were described
(Devi et al., 2010). The authors also investigated the
antibacterial mechanism of membrane permeability by
crystal violet assay, measurement of release of cellular
material absorbed at 260 nm and SDS-PAGE and appli-
cation of AFM (Devi et al., 2010). Nevertheless, the
morphology of S. aureus kept unvaried regardless of
corilagin treatment in the present study.
AFM images can not only be used to obtain qualita-
tive information of biological samples at the nanometer
scale, but can also measure quantitative parameters of
biomaterial properties under nearly physiological condi-
tions (Dvorak, 2003). Therefore, it is easy to compare
modified and unmodified microbial regions. In the
Phytother. Res. 27: 1517–1523 (2013)
1522 N. LI ET AL.

Figure 4. AFM images of E. coli, S. aureus and C. albicans treated with different concentrations of corilagin (control, MIC and 4MIC) for 8 h.
A: E. coli; B: S. aureus; C: C. albicans.

present study, morphological parameters of microorgan-
isms were precisely measured by the NanoScope soft-
ware (Molecular Imaging Inc.). As listed in Table 2,
the mean length, width and height of untreated E. coli
were 2092.21  7.21 nm, 508.32  24.53 nm and 231.16 
36.72 nm, respectively, and turned to 3330.46  27.94 nm,
524.96  6.37 nm and 94.27  37.21 at 4MIC. It experi-
enced a decrease in height which was similar with
that of C. albicans. However, the values of S. aureus
remained very similar. Meanwhile, surface roughness
analysis was performed to describe the structural changes
of cell wall in terms of quantity. The average RMS
roughness of E. coli were calculated as (2.82  0.69 nm),
(5.31  2.16 nm) and (7.48  0.74 nm) for the non-treated,
MIC-treated and 4MIC-treated cells, correspondingly
(Table 2). Similar changes of Rrms calculated from
C. albicans were in accord with the rough and broken
membrane. The slight morphology changes of S. aureus
were also consistent with the value of Rrms. Devi et al.
(2010) pointed out that great alterations of cell morph-
ology observed by AFM could further explain the
antibacterial mechanism of membrane permeability and
resulting in cell growth inhibition. Thus great changes
of morphology, length, width, height and root mean
square values of corilagin-treated E. coli and C. albicans
further confirm the data on increasing membrane
permeability. The slight changes of morphological
parameters and the integrated morphology in Fig. 4-B
showed that corilagin did not change membrane
permeability of S. aureus. This result was in accord with
that of the crystal violet assay.
A synopsis of all our results clearly indicates that corila-
gin affected the membrane permeability of E. coli and C.
albicans but not of S. aureus. Brownlee et al. (1990)
reported that tannins affect the metabolism of microor-
ganisms by modifying the morphology of Crinipellis
perniciosa. It is in line with this point of view, if we
suppose that corilagin might inhibit oxidative

Table 2. Measurement of microorganism dimensions after exposure to corilagin at different concentrations.

E. coli S. aureus C. albicans

p
Concentration Control MIC 4MIC Control MIC 4MIC Control MIC 4MIC Value

Length (nm) 2092.21 2669.14 3330.46 842.23 813.22 750.16 5181.82 6545.51 7091.03 <0.05
7.21 2.49 27.94 12.52 8.53 23.82 5.72 11.48 17.49
Width (nm) 508.32 519.41 524.96 834.74 805.38 736.75 2727.34 2209.17 1963.76
24.53 8.92 6.37 9.05 15.36 17.43 5.94 9.63 7.53
Height (nm) 231.16 160.96 94.27 83.63 75.62 79.48 641.37 492.43 183.94
36.72 24.39 37.21 28.47 19.41 31.28 18.28 27.42 31.26
RMS roughness (nm) 2.82 5.31 7.48 1.32 1.54 1.49 1.23 1.81 5.94
0.69 2.16 0.74 0.27 0.16 0.21 0.14 0.47 0.25

Copyright © 2012 John Wiley & Sons, Ltd. Phytother. Res. 27: 1517–1523 (2013)
 ANTIMICROBIAL ACTIVITY OF CORILAGIN
phosphorylation or the electron transport system by
breaking down the membrane of E. coli and C. albicans.
However, the action mode of corilagin might be more
complicated. Future work should be done to look for spe-
cific targets on the cell surface or the cytoplasm.
In conclusion, corilagin displayed antimicrobial activity
against E. coli, S. aureus and C. albicans, and this was the
first time that its antimicrobial mechanism has been inves-
tigated. As for E. coli, corilagin severely increased the
release of b-lactamase and b-galactosidase activity via
changing OM and IM permeability. However, corilagin
did not act on the protein expression of E. coli as proved
by SDS-PAGE of whole-cell protein lysates. It is worth
speculating that this might be due to its DNA damages ra-
ther than proteins. Meanwhile, corilagin inhibited the
growth of S. aureus by inhibition of protein expression in-
stead of destroying membrane permeability. Moreover,
corilagin changed membrane permeability of C. albicans
and acted on specific proteins. With the help of AFM,
morphological changes of E. coli and C. albicans but not
S. aureus were explicitly proved. To sum up, our results
showed a good consistency on the antimicrobial mechan-
ism of tannins as previously described (Chung et al., 1998;
Cowan, 1999), including changing morphology, increasing
REFERENCES
Askew C, Sellam A, Epp E, et al. 2009. Transcriptional regulation
of carbohydrate metabolism in the human pathogen Candida
albicans. PLoS Pathog 5: e1000612.
Boden Wastfelt MK, Flock JI. 1995. Incidence of the highly
conserved fib gene and expression of the fibrinogen-binding
(Fib) protein among clinical isolates of Staphylococcus aureus.
J Clin Microbiol 33: 2347–2352.
Braga PC, Ricci D. 1998. Atomic force microscopy: application to
investigation of Escherichia coli morphology before and after
exposure to cefodizime. Antimicrob Agents Ch 42: 18–22.
Brownlee HE, McEuen AR, Hedger J, Scott IM. 1990. Antifungal
effects of cocoa tannin on the witches’ broom pathogen
Crinipellis perniciosa. Physiol Mol Plant P 36: 39–48.
Cho J, Lee DG. 2011. The characteristic region of arenicin-1
involved with a bacterial membrane targeting mechanism.
Biochem Bioph Res Co 405: 422–427.
Chung KT, Wong TY, Wei CI, Huang YW, Lin Y. 1998. Tannins and
human health: a review. Crit Rev Food Sci 38: 421–464.
Cowan MM. 1999. Plant Products as Antimicrobial Agents. Clin
Microbiol Rev 12: 564–582.
Demirci F, Guven K, Demirci B, Dadandi MY, Baser KHC. 2008.
Antibacterial activity of two Phlomis essential oils against
food pathogens. Food Control 19: 1159–1164.
Devi KP, Nisha SA, Sakthivel R, Pandian SK. 2010. Eugenol (an
essential oil of clove) acts as an antibacterial agent against
Salmonella typhi by disrupting the cellular membrane. J Ethno-
pharmacol 130: 107–115.
Dvorak JA. 2003. The application of atomic force microscopy to the
study of living vertebrate cells in culture. Methods 29: 86–96.
Fogliani B, Raharivelomanana P, Bianchini JP, Bouraïma-Madjèbi S,
Hnawia E. 2005. Bioactive ellagitannins from Cunonia macro-
phylla, an endemic Cunoniaceae from New Caledonia. Phyto-
chemistry 66: 241–247.
Fu YJ, Zu YG, Chen LY, et al. 2007. Investigation of antibacterial
activity of rosemary essential oil against Propionibacterium acnes
with atomic force microscopy. Planta Med 73: 1275–1280.
Girasole M, Pompeo G, Cricenti A, et al. 2007. Roughness of the
plasma membrane as an independent morphological parameter
to study RBCs: a quantitative atomic force microscopy investi-
gation. Acta Bioch Bioph Sin 1768: 1268–1276.
Haslam E. 1996. Natural polyphenols (vegetable tannins) as drugs:
possible modes of action. J Nat Prod 59: 205–215.
Ibrahim HR, Thomas U, Pellegrini A. 2001. A helix-loop-helix peptide
at the upper lip of the active site cleft of lysozyme confers potent
antimicrobial activity with membrane permeabilization action.
J Biol Chem 23: 43767–43774.
Copyright © 2012 John Wiley & Sons, Ltd.
1523
membrane permeability and destroying proteins of micro-
organisms. However, we believe that the action mode of
corilagin is probably more complex than demonstrated
here. Our future work should aim at clarifying the actual
protein targets on the cell surface or other intracellular
targets. In addition, these results also suggested that cori-
lagin might become a potential antimicrobial for use in
the pharmaceutical industry.
Acknowledgements
The authors gratefully acknowledge the financial supports by Heilongjiang
Province Science Foundation for Excellent Youths (JC200704), Special
Fund of Forestry Industrial Research for Public Welfare of China
(201004040), Program for Importation of International Advanced
Forestry Science and Technology, National Forestry Bureau (2012-4-06)
and Project for Distinguished Teacher Abroad, Chinese Ministry of Edu-
cation (MS2010DBLY031).
Conflict of Interest
The authors have declared that there is no conflict of interest.
Laemmli UK. 1970. Cleavage of structural proteins during the
assembly of the bacteriophage T4. Nature 227: 680–685.
Lehrer RI, Barton A, Ganz T. 1988. Concurrent assessment of inner
and outer membrane permeabilization and bacteriolysis in E. coli
by multiple-wavelength spectrophotometry. J Immunol Methods
108: 153–158.
Li RG, Manna AC, Dai SD, Cheung AL, Zhang GY. 2003. Crystal
Structure of the SarS Protein from Staphylococcus aureus. J
Bacteriol 185: 4219–4225.
Nagaoka I, Hirota S, Yomogida S, Ohwada A, Hirata M. 2000.
Synergistic actions of antibacterial neutrophil defensins and
cathelicidins. Inflamm Res 49: 73–79.
Prasad R, De Wergifosse P, Goffeau A, Balzi E. 1995. Molecular
cloning and characterization of a novel gene of Candida
albicans, CDR1, conferring multiple resistance to drugs and
antifungals. Curr Genet 27: 320–329.
Rajan PA, Richard PN. 2008. Regulatory organization of the
staphylococcal sae locus. Microbiology 154: 949–959.
Scalbert A. 1991. Antimicrobial properties of tannins. Phytochem-
istry 30: 3875–3883.
Spanos A, Sedgwick SG, Yarranton GT, Hübscher U, Banks GR.
1981. Detection of the catalytic activities of DNA
polymerases and their associated exonucleases following
SDS-polyacryla-mide gel electrophoresis. Nucleic Acids Res
9: 1825–1839.
Thennarasu S, Nagaraj R. 1996. Specific antimicrobial and hemolytic
activities of 18-residue peptides derived from the amino terminal
region of the toxin pardaxin. Protein Eng 9: 1219–1224.
Vaara M, Vaara T. 1981. Outer membrane permeability barrier
disruption by Polymixin in Polymixin-susceptible and resist-
ant Salmonella typhimurium. Antimicrob Agents Ch 19:
578–583.
Xing K, Chen XG, Kong M, Liu CS, Cha DS, Park HJ. 2009. Effect of
oleoyl-chitosan nanoparticles as a novel antibacterial dispersion
system on viability, membrane permeability and cell morphology
of Escherichia coli and Staphylococcus aureus. Carbohyd Polym
76: 17–22.
Zhao L, Zhang SL, Tao JY, et al. 2008. Preliminary exploration on
anti-inflammatory mechanism of Corilagin (beta-1-O-galloyl-
3, 6-(R)-hexahydroxydiphenoyl-D-glucose) in vitro. Int Immu-
nopharmacol 8: 1059–1064.
Zu YG, Liu XL, Fu YJ, Wu N, Kong Y, Wink M. 2010. Chemical
composition of the SFE-CO2 extracts from Cajanus cajan (L.)
Huth and their antimicrobial activity in vitro and in vivo. Phyto-
medicine 17: 1095–1101.
Phytother. Res. 27: 1517–1523 (2013)