Colloids and Surfaces B: Biointerfaces 94 (2012) 143150

Contents lists available at SciVerse ScienceDirect

Colloids and Surfaces B: Biointerfaces

journal homepage: www.elsevier.com/locate/colsurfb

Studies on antibacterial activity of ZnO nanoparticles by ROS induced lipid

peroxidation
R.K. Dutta a, , Bhavani P. Nenavathu a , Mahesh K. Gangishetty a , A.V.R. Reddy b
a
Analytical Chemistry Laboratory, Department of Chemistry, Indian Institute of Technology, Roorkee 247667, India
b
Analytical Chemistry Division, Bhabha Atomic Research Centre, Trombay, Mumbai 400085, India

a r t i c l e i n f o a b s t r a c t

Article history: Recent studies indicated the role of ROS toward antibacterial activity. In our study we report ROS medi-
Received 6 September 2011 ated membrane lipid oxidation of Escherichia coli treated with ZnO nanoparticles (NPs) as supported
Received in revised form by detection and spectrophotometric measurement of malondialdehyde (MDA) by TBARS (thiobarbi-
16 December 2011
turic acid-reactive species) assay. The antibacterial effects of ZnO NPs were studied by measuring the
Accepted 22 January 2012
growth curve of E. coli, which showed concentration dependent bacteriostatic and bacteriocidal effects
Available online 7 February 2012
of ZnO NPs. The antibacterial effects were characterized by scanning electron microscopy (SEM) and
transmission electron microscopy (TEM). Further, antibacterial effect of ZnO NPs was found to decrease
Keywords:
ZnO nanoparticles
by introducing histidine to the culture medium treated with ZnO NPs. The ROS scavenging action of
Reactive oxygen species histidine was conrmed by treating histidine to the batch of Escherichia coli + ZnO NPs at the end of the
Lipid peroxidation lag phase of the growth curve (Set-I) and during inoculation (Set-II). A moderate bacteriostatic effect
Malondialdehyde (lag in the E. coli growth) was observed in Set-II batch while Set-I showed no bacteriostatic effect. From
Antioxidant these evidences we conrmed that the antibacterial effect of bare as well as TG capped ZnO NPs were
Electron microscopy due to membrane lipid peroxidation caused by the ROS generated during ZnO NPs interaction in culture
medium.
2012 Elsevier B.V. All rights reserved.

1. Introduction
Metals and metal oxide nanomaterials e.g., Ag, TiO2 , SiO2 , MgO
and CaO are getting signicant scientic attention due to their vast
existing as well as potential newer applications, especially as bio-
cidal or disinfecting agents [14]. In this regard, ZnO nanoparticles
(NPs) were found to be versatile and are considered to be poten-
tial next generation materials as biocidal or disinfecting agent,
which is attributed to their higher stability than the organic based
disinfectants or anti-microbial agents. Lately, ZnO NPs were stud-
ied for their antibacterial [511] and anticancer properties [12].
Various mechanistic approaches were proposed to explain the
anti-bacterial action of ZnO NPs, including their tentative role in
generating reactive oxygen species (ROS), release of zinc ions,
cellular membrane dysfunction and nanoparticle internalization.
Reported evidences have indicated that ROS might be the most
important cause for antibacterial effect [13]. In this regard, Sawai
et al. [14] made a signicant contribution by demonstrating pro-
duction of H2 O2 in the aqueous suspension of ZnO NPs. Further,
Lipovsky et al. [8] detected hydroxyl radicals and singlet oxygen
Corresponding author. Tel.: +91 1332 285280; fax: +91 1332 286206.
E-mail address: duttafcy@iitr.ernet.in (R.K. Dutta).
in the aqueous suspensions of ZnO NPs by EPR method and also
demonstrated enhanced antibacterial activity of nanocrystalline
ZnO attributed due to increased ROS-mediated cell injury [15].
In a recent study they have elucidated the role of ROS mediated
antifungal effect of ZnO NPs where ROS levels determined by EPR,
decreased in the presence of histidine [16].
We have carried out a systematic study to elucidate the role
of ROS toward antibacterial activity of ZnO NPs. In a biological
system, ROS is determined by TBARS method which is based on
formation of a pink colored complex between thiobarbituric acid
and malondialdehyde (MDA) that can be quantitatively measured
spectrophotometric method at max = 532 nm [17]. The MDA is pro-
duced during the lipid peroxidation of unsaturated fatty acid by
oxygen based free radicals and the measured intensity is related to
the concentration of generated ROS. The TBARS assay was chosen
to screen and monitor the mechanism of lipid peroxidation which
is a major indicator of oxidative stress [18,19]. Besides, it is rapid,
easy to use, inexpensive and sensitive method. Further, this assay
is reported to be useful for measuring antioxidant activity of var-
ious compounds [20,21]. We have determined MDA levels which
were the measures of ROS generated in the batches of culture media
treated with increasing concentrations of ZnO NPs (bare and thio-
glycerol capped). It has been discussed that the ROS generated in
the culture media might be due to oxidation of membrane lipids

0927-7765/$ see front matter 2012 Elsevier B.V. All rights reserved.
doi:10.1016/j.colsurfb.2012.01.046
144 R.K. Dutta et al. / Colloids and Surfaces B: Biointerfaces 94 (2012) 143150
of E. coli leading to their membrane dysfunction and consequently
leading to bacteriostatic or bactericidal conditions. In order to prove
the role of ROS, we further demonstrated reduction of antibacterial
activity of ZnO NPs in the presence of histidine via TBARS studies.
2. Experimental
2.1. Materials
The wild type Escherichia coli (E. coli K-12 strain MTCC 1302) was
obtained from IMTECH, Chandigarh, India. Thioglycerol (TG), ZnCl2 ,
methanol, NaOH, sodium dodecyl sulfate (SDS) and Luria-Bertani
(LB) broth media were procured from Himedia, India. Thiobarbitric
acid, linolenic acid, histidine, 1,1,3,3 tetramethoxypropane (99%
pure) and commercial grade (bare) ZnO nanoparticles (<100 nm
in size, 99.99%), were procured from Sigma Aldrich, Germany. All
chemicals and solvents were used without further purication.
2.2. Synthesis of TG capped ZnO NPs
The TG capped ZnO NPs were synthesized by wet-chemical
method using NaOH and ZnCl2 as precursors in methanol and the
pH 7.2 was maintained during the synthesis. 50 mL of 20 mM of
ZnCl2 solution was prepared in methanol, to which 8.5 L of TG was
added and stirred for 1 h so that the nal concentration of TG in the
solution was 2 mM. About 0.4 g of NaOH pellets were added directly
to the above mix and the entire solution was stirred for 4 h. The ZnO
NPs were collected after precipitating with n-hexane. The excess TG
was drained out and the NPs after washing with ethanol were kept
for overnight drying at 60 C. The TG was used as a stabilizing agent
for controlling the size of the ZnO NPs by preventing agglomeration.
Powdered nanoparticles were obtained after washing and dry-
ing the precipitate. These nanoparticles were characterized after
re-dispersing in methanol. The purpose of studying antibacterial
effects of TG capped ZnO NPs was to evaluate whether the capping
agent has any role on antibacterial effect of ZnO NPs. In addition,
applying capping agent usually results in smaller sized particles
so it would be possible to assess the size dependent antibacterial
effect as well. In this regard, it should be mentioned that the con-
centration of TG used in the synthesis of TG capped ZnO NPs did
not show antibacterial activity as the concentration of TG used was
much less than the reported MIC [22].
2.3. Characterization
The X-ray diffraction measurements of commercial ZnO NPs
and TG capped ZnO NPs were performed with powder diffrac-
tometer (Bruker ARS D8 Advance) operated at 40 kV using graphite
monochromatized Cu K radiation source with a wavelength of
1.54 A in a wide-angle region from 20 to 80 on a 2 scale. The opti-
cal properties were measured by UVvisible spectrophotometer
(Shimadzu, UV-1800) in the range of 200500 nm. The particle size
distribution and morphology of bare (commercial) ZnO nanoparti-
cles and TG capped ZnO NPs were studied using FESEM EDAX (eld
emission scanning electron microscope coupled to energy disper-
sive X-ray analyzer), FEI-Quanta 200F operated at 20 kV. The sample
preparation for SEM measurement was done by spraying the dis-
persion of ZnO nanomaterials on a clean glass plate, dried at room
temperature and coated with a thin layer of Au for electrical con-
ductivity for incident electrons. The interaction of nanoparticles of
ZnO with E. coli was imaged by transmission electron microscopy
(TEM) using FEI Technai-G2 microscope operated at 200 kV. The
samples for TEM analysis were collected at the beginning of station-
ary phase and prepared by placing a drop of the culture medium
containing E. coli treated with ZnO nanoparticles (after suitable
dilution with buffer solution) on a carbon coated 150 mesh copper
grid and dried at room temperature. The thermogravimetric analy-
sis of TG capped ZnO nanomaterials was carried out with a heating
rate of 5 C min1 from ambient temperature up to 1000 C under
a nitrogen ow (200 mL min1 ) using Perkin Elmer Pyris Diamond,
to ensure weight loss in the sample due to thermal degradation of
polymer.
2.4. Antibacterial activity
The bare as well as TG capped ZnO NPs were assayed for
antibacterial activity using growth inhibition studies against gram-
negative bacteria E. coli strain K-12. For antibacterial assay, the
bacterial colonies were rst grown on solid nutrient agar medium
and from the primary cultured agar plates, fresh colonies were
inoculated into 100 mL nutrient broth medium comprising of Luria-
Bertani (LB) media containing 5 g L1 of yeast extract, 10 g L1 of
casein enzymic hydrolysate and 10 g L1 of NaCl in water. The E. coli
growth in the culture media was monitored by measuring its opti-
cal density (O.D) at 600 nm. When the O.D reached to mid log phase
(i.e., O.D = 0.6 units), a 10 L of this culture was retrieved and was
inoculated in 100 mL of freshly prepared nutrient LB broth medium
supplemented with different concentrations of bare ZnO NPs of
concentration ranging between 8 and 45 mg/100 mL of media. Sim-
ilar process was followed for studying the batches treated with
TG-capped ZnO NPs of concentration ranging between 8 and 55 mg
per 100 mL media. Positive controls consisting of NPs in LB medium
without inoculum and negative controls consisting of LB medium
with only inoculum were used. All the asks were then incubated
at 37 C in an orbital shaker at 150 rpm. The bacterial growth was
monitored at an interval of 1 h by measuring the O.D of the culture
media at  = 600 nm.
2.5. Determination of reactive oxygen species (ROS)
The ROS generated in the culture media was determined using
TBARS assay. An aliquot of 1 mL of culture media was collected
to which 200 L of 10% SDS was added and swirled vigorously.
2 mL of freshly prepared thiobarbituric acid (TBA) was added to
the above mix and was incubated at 95 C for 60 min. It was
cooled to room temperature and was centrifuged at 3500 rpm for
15 min. The absorption spectrum of the supernatant was recorded
at max = 532 nm to estimate the formation of TBARS. All anal-
yses were carried out in triplicate. From the TBARS intensities
the corresponding levels of MDA were deduced from a calibra-
tion curve plotted between the intensity of TBARS measured at
 = 532 nm against different concentrations of MDA synthesized
by acidic hydrolysis of 1,1,3,3 tetramethoxypropane (Supporting
information, Fig. S1).
2.6. Study of antibacterial effect of ZnO NPs treated E. coli in the
presence of antioxidants
Two sets of experiments were carried out to assess the effect
of an antioxidant, e.g., histidine as a scavenger of ROS and con-
sequently reduction of antibacterial activities of ZnO NPs. The
growth patterns of E. coli treated with commercial ZnO NPs of
35 mg/100 mL media) and TG capped ZnO NPs (55 mg/100 mL
broth) were studied in the presence of histidine. The reason for
choosing these concentrations of bare ZnO NPs and TG capped ones
was both conditions exhibited similar bacteriostatic growth curves
hence expected to generate similar levels of ROS.
Set-I: the methodology for cell culture was the same as dis-
cussed above i.e., 10 L of K-12 E. coli strain was inoculated in
freshly prepared nutrient broth, to which histidine of concentration
ranging between 15 mg and 45 mg/100 mL of culture media was
introduced after 3 h from the addition of inoculum, corresponding
 R.K. Dutta et al. / Colloids and Surfaces B: Biointerfaces 94 (2012) 143150
to the end of the lag phase of the growth curve. The delayed addition
of histidine was to ensure the ROS generation.
Set-II: histidine concentration of 45 mg/100 mL was added at
the time of addition of inoculum E. coli + ZnO NPs (35 mg/100 mL
media) to conrm the role of ROS in the antibacterial action of ZnO
NPs. Respective growth curves and the MDA levels were measured
for both the sets of experiments.
3. Results and discussion
The bare ZnO NPs (commercially procured) showed X-ray
diffraction peaks at (1 0 0), (1 0 1), (1 0 2), (1 1 0), (1 0 3) and (1 1 2)
planes (Supporting information, Fig. S2) corresponded to hexagonal
wurtzite phase of ZnO, JCPDS card No: 5-0664. The crystallite size
(D) was calculated from Debye Scherrer formula D = 0.9/ cos();
where  is the wavelength of X-ray source, is full-width at half
maximum (FWHM) in radians and  is Braggs diffraction angle.
The morphology and size distribution of the bare ZnO NPs were
also conrmed from FESEM studies where the energy dispersive
X-ray analysis (EDAX) of these NPs exhibited characteristic X-ray
peaks of Zn and O (Fig. S3a). The size of these particles appeared to
be less than 100 nm and was in good agreement with the reported
average size by Sigma Aldrich.
The thioglycerol (TG) capped ZnO NPs were successfully syn-
thesized as evidenced from its XRD pattern, which matched with
those of bare ZnO NPs (Supporting information, Fig. S2). The peaks
were broad and the crystallite size (D) of these TG capped nanopar-
ticles was measured to be 4 nm. The broad diffraction peaks
observed for TG capped ZnO NPs could be due to various reasons,
namely, small particle sizes together with strain due to defects in
the nanomaterials. Zak et al. [23] used modeled study to explain
the broadening of XRD peak of ZnO NPs. The scanning electron
microscopy study of the TG capped ZnO nanoparticles showed sig-
natures of agglomeration of spherical shaped nanoparticles and
their EDAX analysis exhibited the characteristic K and L X-rays of
zinc, K X-ray of oxygen (Supporting information, Fig. S3b).
The TG capped ZnO NPs showed a blue shift with excitonic peak
measured at 310 nm (Supporting information, Fig. S4a) as com-
pared to the UVvisible absorption band measured at 370 nm for
commercial ZnO NPs (Fig. S4b). The corresponding band gap for
TG capped ZnO NPs was calculated to be 4.0 eV, which is typically
due to particles of sizes less than 5 nm of ZnO NPs. The absorp-
tion peak was quite sharp, indicating monodispersed nature of the
ZnO NPs. The TG capping on ZnO NPs were further conrmed from
thermogravimetric analysis which showed 23% mass loss due to
degradation of thioglycerol capping and also removal of absorbed
water (Supporting information, Fig. S5) and was in good agreement
with the reported literature [24]. From thermal analysis, it was
deduced that the thioglycerol concentration in the TG-capped NPs
was 0.46 mM, which was much lower than the reported minimum
inhibitory concentration (2 mM) of TG for E. coli growth.
3.1. Evaluation of antibacterial effect of ZnO NPs
The growth patterns of the K-12 strain of E. coli treated with
the bare (commercial) ZnO NPs of concentrations ranging between
8 mg and 45 mg in 100 mL of LB media were assayed (Fig. 1a). The
growth curve of the batch treated with 8 mg of bare ZnO NPs was
similar to that of the control batch. The minimum inhibitory con-
centration (MIC) for bare ZnO NPs was found to be 15 mg/100 mL
of culture media. The inhibition of bacterial growth was found to
increase with the concentrations of ZnO NPs treated in culture
media. The antibacterial effect of these bare ZnO NPs was also evi-
dent from the delay in the onset of the logarithmic growth phase,
referred to as bacteriostatic condition. The delay was found to be
145
more for higher concentrations of ZnO NPs. Signicant delay (2 h)
in the onset of the growth was observed for the batch treated with
35 mg of bare ZnO NPs/100 mL culture media and its corresponding
inhibition of growth was estimated to be 80% with respect to the
control batch. Complete inhibition of the bacterial growth, i.e., bac-
teriocidal condition was observed for the batch treated with 45 mg
of bare ZnO NPs/100 mL media, corresponding to minimum bac-
tericidal concentration. It may be surmised that the antibacterial
effect of ZnO NPs were dependent on the concentration ZnO NPs
treated in the culture media and was in good agreement with the
previous reports [411].
Similarly, the growth patterns of the K-12 strain of E. coli treated
with the TG capped ZnO NPs of concentrations ranging between
8 mg and 55 mg in 100 mL of LB media were assayed (Fig. 1b). The
batches treated with 8 mg and 15 mg of the TG capped ZnO NPs
were similar to that of the control while the minimum inhibitory
concentration for the capped NPs was found to be 25 mg/100 mL
culture media. The TG capped ZnO NPs also showed bacteriostatic
condition at higher concentrations of treated TG capped ZnO NPs.
Unlike bare ZnO NPs, the bacteriocidal condition for TG capped ZnO
NPs was not observed even at a concentration of 55 mg/100 mL
culture media. The batch treated with 55 mg of TG capped ZnO
NPs/100 mL media exhibited 80% of inhibition of the growth with
respect to the control batch. Interestingly the inhibition pattern
was similar to the batch of 35 mg of bare ZnO NPs/100 mL.
From these studies, it was evident that the bare ZnO NPs showed
higher antibacterial activity than the TG capped ones. Notably, the
bare ZnO NPs were several folds bigger in size than the TG capped
ZnO NPs. Therefore our studies contradicted to the earlier reports
that the smaller sized ZnO NPs exhibit more antibacterial effect
[11]. It is therefore interesting to study the mechanism underly-
ing the antibacterial effect due to both capped as well as bare ZnO
NPs. In this regard, several mechanisms have been proposed by
different researchers, which can be broadly classied into three
categories [13]: (a) physical processes, comprising of adsorption
of ZnO NPs on the surface of the E. coli cells; (b) biological pro-
cesses leading to internalization of NPs in the E. coli, via channels
or proteins at the cell wall and (c) chemical phenomena result-
ing in production of reactive species. Further, antibacterial effect
of ZnO NPs due to combination of these phenomena could not
however be ruled out. In order to understand the antibacterial
effect, it is important to realize the structure of an E. coli (gram
negative bacteria). It comprises of cell wall with three layered mem-
brane structure as shown in a schematic diagram (Fig. 2). The outer
layer consists of membrane proteins, porins, phospholipids and
lipopolysaccharides. The middle layer consists of peptidoglycan.
The inner cytoplasmic membrane consists of 40% phospholipids
and 60% protein. The gram negative cell wall of E. coli is due to nega-
tively charged lipopolysaccharides which might allow the ZnO NPs
to interact electrostatically [25]. A major functional consequence of
interaction of these ZnO NPs could lead to breakdown of the mem-
brane barrier of E. coli, which could accompany dissipation of ion
gradients and might eventually lead to cell swelling and lysis [26].
In addition, ZnO NPs are reported to be internalized via the usual
transport channels and the internalized NPs might interact with the
cellular constituents to cause antibacterial effect. Brayner et al. [5]
Huang et al. [7] and Nair et al. [12] have demonstrated using TEM
data that ZnO NPs of sizes 11 nm and 60 nm were internalized in
E. coli and corroborated these results with alteration of morphology
and damage of bacterial cell. However, the mode of internaliza-
tion of such large particles were non-trivial as for gram-negative
prokaryotic cells, e.g., E. coli, the average maximum size range for
non-specic uptake of large molecules is of the order of 1.82 nm
via an OmpF porins [27,28]. Our TEM studies showed adherence
of bare ZnO NPs (Fig. 3a) and TG capped ZnO NPs (Fig. 3b) on the
surface of E. coli. The adherence of NPs on the surface of the E. coli
146 R.K. Dutta et al. / Colloids and Surfaces B: Biointerfaces 94 (2012) 143150

(a) Control
8mg of ZnO NPs per 100mL CM
(b)
Control
15mg of ZnO NPs per 100mL CM 15mg of TG-ZnO NPs per 100mL CM
25mg of ZnO NPs per 100mL CM 25mg of TG-ZnO NPs per 100mL CM
30mg of ZnO NPs per 100mL CM
45mg of TG-ZnO NPs per 100mL CM
35mg of ZnO NPs per 100mL CM
55mg of TG-ZnO NPs per 100mL CM
45mg of ZnO NPs per 100mL CM
2.0
2.0
1.6
1.6

O.D 600 nm
O.D600 nm

1.2 1.2

0.8 0.8

0.4 0.4

0.0
0.0
0 2 4 6 8 10 12 0 4 8 12
Time (h) Time (h)

Fig. 1. Growth patterns of E. coli treated with different concentrations of (a) bare ZnO NPs; (b) TG capped ZnO NPs (CM, culture media). The results correspond to mean SD
(SD, standard deviation; n = 3).

might lead to physical blockages of the transport channels of the cell

membranes, due to which transportation of cellular nutrients from
media were restricted, resulting into cell starvation and eventually
cell mortality. Deformation of cellular morphology, cell swelling
and membrane rupture were observed in our SEM study (Fig. 3c)
showing ZnO particles on the surface of the cell wall by EDAX anal-
ysis. Further, TEM study of ZnO NPs treated with E. coli also revealed
deformation of cellular morphology, cell swelling and membrane
rupture (Fig. 3d). It may be remarked here that the damages of cell
membrane might have occurred due to oxidation of lipid mem-
brane in the cell wall which resulted in porous cell membrane and
facilitated internalization of ZnO NPs in cells.
The oxidation of membrane lipids could be triggered by reactive
oxygen species like H2 O2 [4] or highly reactive hydroxyl radical
and singlet oxygen [15], which were reported to be generated in
ZnO nanoparticles suspended in water. The mechanism for the pro-
duction of ROS (H2 O2 and OH ) was attributed to light induced
effect. Notably, ordinary room light with a total light intensity of
10 W/cm1 was sufcient for generating ROS [15]. The mecha-
nism of light induced generation of ROS could be given as follows
[29],
ZnO + h ZnO + e + h+ ; h+ + H2 O OH + H+
e + O2 O2
O2 + H+ HO2
HO2 + e + H+ H2 O2
The hydroxyl radicals generated in suspensions of ZnO NPs in
culture media could extract the hydrogen atom from allylic position
of the unsaturated fatty acid. The allyl free radical that is formed
might react with the oxygen molecule to form lipid peroxide rad-
ical (LOO ), which could undergo rearrangement to form MDA, as
schematically shown in (Supporting information, Fig. S6). The MDA
binds with thiobarbitiric acid to form a complex referred to as
TBARS which corresponds to a characteristic UVvisible absorption
peak at max = 532 nm and hence can be determined by spectropho-
tometric method. The intensity of the absorption at max = 532 nm
is directly related to the amount of MDA formed, which is equiv-
alent of ROS generated during the interaction of ZnO NPs in the
culture media.
Compared to the control batch (i.e., without ZnO NPs in culture
media), the MDA equivalents for the bare as well as TG capped
ZnO NPs treated batches were higher (Fig. 4a) and increased lin-
early (R2 = 0.94) with increasing concentrations of bare ZnO NPs
and TG capped ZnO NPs (Fig. 4b). This indicated increase in the ROS
generation due to higher concentrations of ZnO NPs in the culture
media. It was noted that the MDA equivalent (ROS equivalence)
was about 36% higher in the batch treated with 45 mg of bare ZnO
NPs in 100 mL of culture media (which corresponded to 100% inhi-
bition of E. coli growth, i.e., bactericidal) as compared to the control
batch. The batches treated with 35 mg, 25 mg and 15 mg of bare
ZnO NPs per 100 mL media, which showed bacteriostatic condition,

Porins Lipopolysachcharides

Membrane Phospholipid
Protein

Outer membrane
ZnONPs C (8nm)

Peptidoglycan

Cytoplasmic
Cytoplasm membrane

Fig. 2. Schematic structure of E. coli showing potential interaction with ZnO NPs.
 R.K. Dutta et al. / Colloids and Surfaces B: Biointerfaces 94 (2012) 143150 147

Fig. 3. (a) TEM images showing adherence of bare ZnO NPs on the surface of E. coli, (b) TEM images of TG capped ZnO NPs treated E. coli, (c) SEM image showing deformed
E. coli engulfed by bare ZnO NPs, supported by EDAX spectrum showing Zn L-X-ray, and (d) TEM image of E. coli showing cell swelling and cell wall deformation due to
interaction of ZnO NPs.

corresponded to generation of about 23.6%, 16% and 5.5% of MDA
equivalent respectively. Similarly, with respect to the control batch,
the relative amounts of MDA equivalent in the batches treated with
25 mg, 45 mg and 55 mg of TG capped ZnO NPs were 6.9%, 13.3%
and 24%, respectively. The measured MDA equivalent, which was
equivalent to the generation of ROS, correlated with the growth
patterns of bare and TG capped ZnO NPs treated E. coli. From these
observation it may be remarked that the inhibition of the growth
curve due to the increase in the ZnO NPs were attributable to ROS
formation as evident from the increasing order of equivalent MDA
in the culture medium. Our results were in good agreement with
a recent report on toxicity of ZnO nanoparticles on gram positive
bacteria and cancer cells by apoptosis by lipid peroxidation [30].
3.2. Histidine inhibits the antibacterial effect caused by ZnO NPs
In order to conrm the ROS mediated antibacterial effect of
ZnO NPs the growth patterns of E. coli treated with ZnO NPs in
148 R.K. Dutta et al. / Colloids and Surfaces B: Biointerfaces 94 (2012) 143150

(a) Bare ZnO NPs (b) Bare ZnO NPs

40 40

MDA equivalents (nmol/mL)

TG-ZnO NPs
TG-ZnO NPs

MDA (nmol/mL)
35
35

30

30
25

25
0 15 25 35 45 -- 0 10 20 30 40 50
Concentration of ZnO NPs Concentration of ZnO NPs
(mg/100mL) (mg/100mL)

Fig. 4. Showing increase in the MDA equivalents (nmol/mL) with increase in the concentrations of (a) bare ZnO NPs/100 mL of culture media and TG capped ZnO NPs/100 mL
culture media and (b) linear correlation between MDA formed and concentrations of bare and TG capped ZnO NPs in the culture media. The results correspond to mean SD
(SD, standard deviation; n = 3).

Control
45mg Histidine
(a) 25mg Histidine (b)
15mg Histidine
MDA equivalents (nmol/mL)
50
2.0 Without Histidine
1.6 45
bare ZnO NPs
1.2 40
O.D600nm

0.8 35

0.4
30
0.0
0 2 4 6 8 10 12 0 15 25 45 -- --
Time (h) Concentration of Histidine
(mg/100mL)

Control
45mg Histidine
(c) 35mg Histidine (d)
15mg Histidine 55
MDA equivalents (nmol/mL)

2.0
Without Histidine
50
1.5 TG capped ZnO NPs
45
O.D600 nm

1.0 40

35
0.5
30
0.0
0 4 8 12 0 15 25 45 -- --
Time (h) Concentration of Histidine
(mg/100mL)

Fig. 5. Growth curve of batches of E. coli treated with 35 mg of bare ZnO NPs/100 mL media and supplemented with different concentrations of histidine added at the end of
the lag phase of the growth prole, (b) showing decrease in MDA equivalents (nmol/mL) with the increase in histidine concentrations, (c) growth curve of batches of E. coli
treated with 55 mg TG capped ZnO NPs/100 mL media and supplemented with different concentrations of histidine at the end of lag phase of the growth prole, and (d)
showing decrease in MDA equivalents (nmol/mL) with the increase in histidine concentrations. The values for each measurement correspond to mean SD (SD, standard
deviation, n = 3).
 R.K. Dutta et al. / Colloids and Surfaces B: Biointerfaces 94 (2012) 143150 149

the presence of antioxidant were studied. It was thought that the Control
ROS generated due to the interaction ZnO NPs with E. coli could 45mg Histidine (Set II)
be scavenged by the antioxidants. In this regard, we studied the
2.0 45mg Histidine (Set I)
effect of histidine, which is a well known scavenger of hydroxyl
radicals and singlet oxygen [15], toward inhibition of antibacterial
activity of ZnO nanoparticles. Firstly, it was important to conrm 1.6
that histidine did not exhibit antibacterial activity. In order to ver-
ify this, control experiments were carried out where the culture
1.2

O.D600nm
media without ZnO NPs, were treated with 45 mg/100 mL of histi-
dine and compared with the batch without hisitidine (as given in
Fig. S7). It was observed that the growth curve of E. coli for both the 0.8
cases were similar, which conrmed that histidine does not exhibit
antibacterial activity. 0.4
Furthermore, the inhibition of the bacterial growths which were
treated with 35 mg/100 mL ZnO NPs, were reduced as the histidine
concentration was increased from 15 mg/100 mL to 45 mg/100 mL 0.0
of culture medium (Fig. 5a). This was substantiated by decrease
0 2 4 6 8 10 12
in the MDA equivalents with the increase in the histidine con-
centration (Fig. 5b). Similar activity of histidine toward inhibition Time (h)
of antibacterial effect of TG capped ZnO NPs and correspond- Fig. 6. Growth curve of the batch of E. coli treated with 35 mg of bare ZnO
ing decrease in ROS formation was also observed (Fig. 5c and d, NPs/100 mL culture media and supplemented with histidine at the end of the lag
respectively). The decrease in MDA equivalents due to addition of phase, i.e., after 3 h (Set-I) and at the time of addition of inoculum (Set-II). The values
increasing concentration of histidine, indicated scavenging of ROS for each measurement correspond to mean SD (SD, standard deviation, n = 3).
generated during interaction of ZnO NPs in culture media.
To further ascertain the effect of histidine toward scavenging
culture media, and bacteriocidal (no growth of E. coli) condition
ROS generated during interaction of ZnO NPs in culture media,
for the batch treated with 45 mg/100 mL of culture media. While
two sets of experiments were carried out on batches treated with
for TG capped ZnO NPs, bacteriostatic condition prevailed up to
35 mg/100 mL media where 45 mg of histidine was added to the
NP concentration of 55 mg/100 mL which corresponded to about
culture media after 3 h of inoculum corresponding to the end of
80% inhibition of growth and was similar to the batch treated with
the lag phase of the E. coli growth curve (Set-I) and at the time
35 mg/100 mL of bare ZnO NPs. The lesser antibacterial effect of TG
of inoculation (Set-II). The delay in supplementing histidine was
capped NPs was attributed to the capping agent. Further, the role
done to facilitate the production of ROS in the culture media so that
of ROS toward their antibacterial effect of ZnO NPs was conrmed
the generated ROS can be scavenged by the supplemented antiox-
by observing a reduction in the inhibition of bacterial growth due
idants. In the Set-I, the delay in the onset of logarithmic phase of
to external supplementation of histidine in the culture media. The
the growth curve was diminutive. The inhibition of the growth was
corresponding MDA levels were also reduced. This was attributed to
only 10% of the control batch as compared to the 80% inhibition
ROS scavenging action of histidine. These results strongly suggested
observed in the case where histidine was not supplemented (as
that the antibacterial effect of ZnO NPs were due to membrane lipid
shown in Fig. 5a and c). Such signicant reduction of antibacterial
oxidation caused by ROS generated during ZnO NPs interaction with
effect of ZnO NPs could be attributed to ROS scavenging action of
E. coli. The outcome the membrane lipid oxidation perhaps led to
histidine. On the other hand, the batch treated with 35 mg/100 mL
cell wall dysfunctioning and rupture.
of ZnO NPs and histdine together (Set-II), exhibited similar growth
curve to that of the control samples (Fig. 6). Though at such con-
centration of ZnO NPs, the ROS formation was expected, but the Acknowledgments
presence of histidine was sufcient to scavenge the generated ROS.
It was noted from Fig. 5b and d, that histidine was not capable to This work has been carried out under the BRNS project
scavenge all the generated ROS in the culture media. This could be 2007/37/47/BRNS and authors thank BRNS for the support. Mr.
attributed due to its scavenging action of only hydroxyl radical and Mahesh K Gangishetty is thankful to BRNS, India for awarding fel-
singlet oxygen [15], while hydrogen peroxide which is a byproduct lowship. Mr. N. Bhavani Prasad is thankful to MHRD, Govt of India
of cellular metabolism might not be scavenged by histidine. Overall, for awarding senior research fellowship. We would like to thank
accumulated evidences strongly supported the role of ROS toward Mr. Khushwant Singh, one of our research group members and Dr.
antibacterial activity which was most likely due to membrane lipid R. Acharya, Bhabha Atomic Research Centre, Mumbai for their keen
peroxidation. interest in this work. We thank Institute Instrumentation Centre of
IIT Roorkee for allowing us to use the instrumental facilities.

4. Conclusions Appendix A. Supplementary data

Production of reactive oxygen species (ROS) during interaction
of ZnO NPs with E. coli was conrmed to be the key phenomenon
for antibacterial effect of ZnO NPs. The generated ROS in the cul-
ture media was capable to cause oxidation of lipid membrane in
the cell wall of E. coli, as evident from the determination of mal-
ondialdehyde, which is a by-product of lipid peroxidation reaction
of unsaturated fatty acid. The malondialdehyde levels were higher
for the batches treated with higher concentrations of ZnO NPs, and
it was correlated with bacteriostatic conditions for the batches
treated with ZnO NPs of concentration up to 35 mg/100 mL of
Supplementary data associated with this article can be found in
the online version, at doi:10.1016/j.colsurfb.2012.01.046.
References
[1] J.J. Antony, P. Sivalingam, D. Siva, S. Kamalakkannan, K. Anbarasu, R. Sukirtha,
M. Krishnan, S. Achiraman, Colloids Surf. B 88 (2011) 134140.
[2] L. Shtykova, C. Fant, P. Handa, A. Larsson, K. Berntsson, H. Blanck, R. Simonsson,
M. Nydn, H.I. Hrelind, Prog. Org. Coat. 64 (2009) 2026.
[3] S. Makhluf, R. Dror, Y. Nitzan, Y. Abramovich, R. Jelinek, A. Gedanken, Adv. Funct.
Mater. 15 (2005) 17081715.
150 R.K. Dutta et al. / Colloids and Surfaces B: Biointerfaces 94 (2012) 143150
[4] O. Yamamoto, M. Komatsu, J. Sawai, Z. Nakagawa, J. Mater. Sci. Mater. Med. 15
(2004) 847851.
[5] R. Brayner, R. FerrariIliou, N. Brivois, S. Djediat, M.F. Benedetti, Fievet, Nano
Lett. 6 (2006) 866870.
[6] K.M. Reddy, K. Feris, J. Bell, D.G. Wingett, C. Hanley, A. Punnoose, Appl. Phys.
Lett. 90 (2007) 21390212139023.
[7] Z. Huang, X. Zheng, D. Yan, G. Yin, X. Liao, Y. Kang, Y. Yao, D. Huang, B. Hao,
Langmuir 24 (2008) 41404144.
[8] A. Lipovsky, Z. Tzitrinovich, H. Friedmann, G. Applerot, A. Gedanken, R. Lubart,
J. Phys. Chem. C 113 (2009) 1599716001.
[9] R.K. Dutta, P.K. Sharma, R. Bhargava, N. Kumar, A.C. Pandey, J. Phys. Chem. B
114 (2010) 55945599.
[10] R. Wahab, A. Mishra, S.I. Yun, Y.S. Kim, H.S. Shin, Appl. Microbiol. Biotechnol.
87 (2010) 19171925.
[11] K.R. Raghupathi, R.T. Koodali, A.C. Manna, Langmuir 27 (2011) 40204028.
[12] S.K. Nair, A. Sasidharan, V.V. Divya Rani, D. Menon, S. Nair, K. Manzoor, S. Raina,
J. Mater. Sci. Mater. Med. 20 (2009) 235241.
[13] L. Zhang, Y. Jiang, Y. Ding, N. Daskalakis, L. Jeuken, M. Povey, A.J. ONeill, D.W.
York, J. Nanopart. Res. 12 (2009) 16251636.
[14] J. Sawai, E. Kawada, F. Kanou, H. Igarashi, A. Hashimoto, T. Kokugan, M. Shimizu,
J. Chem. Eng. 29 (1996) 627633.
[15] G. Applerot, A. Lipovsky, R. Dror, N. Perkas, Y. Nitzan, R. Lubart, A. Gedanken,
Adv. Funct. Mater. 19 (2009) 842852.
[16] A. Lipovsky, Y. Nitzan, A. Gedanken, R. Lubart, Nanotechnology 22 (2011)
105101105107.
[17] L.T. Rael, G.W. Thomas, M.L. Craun, C.G. Curtis, R. Bar-Or, D. Bar-Or., J. Biochem.
Mol. Biol. 37 (2004) 749752.
[18] K. Yagi, Free Radical Antioxid. Protoc. 108 (1998) 101106.
[19] G. Lefevre, Ann. Biol. Clin. 56 (1998) 305319.
[20] C. Villa, Gac. Med. Mex. 136 (2000) 249256.
[21] A. Hunnisett, Biol. Trace Elem. Res. 47 (1995) 125132.
[22] K.K. Jensen, G.T. Javor, Chemotherapy 24 (1981) 860867.
[23] A.K. Zak, W.H.A. Majid, M.E. Abrishami, R. Youse, Solid State Sci. 13 (2011)
251256.
[24] A.K. Singh, V. Vishwanth, V.C. Janu, J. Lumin. 129 (2008) 874878.
[25] K. Feris, C. Otto, J. Tinker, D. Wingett, A. Punnoose, A. Thurber, M.
Kongara, M. Sabetian, B. Quinn, C. Hanna, D. Pink, Langmuir 26 (2010)
44294436.
[26] R.W. Titball, A.K. Basak, in: J.E. Alouf, M.R. Popoff (Eds.), A Comprehensive
Sourcebook of Bacterial Protein Toxins, 3rd Ed., Academic Press, London, 2006,
pp. 516534.
[27] R. Koebnik, K.P. Locher, P.V. Gelder, Mol. Microbiol. 37 (2000) 239259.
[28] G.E. Schulz, Curr. Opin. Struct. Biol. 10 (2000) 443447.
[29] N. Padmavathy, R. Vijayaraghavan, Sci. Technol. Adv. Mater. 9 (2008) 37.
[30] M. Premanathan, K. Karthikeyan, K. Jeyasubramanian, G. Manivannan,
Nanomedicine 7 (2011) 184192.