Colloids and Surfaces B: Biointerfaces 135 (2015) 316323

Contents lists available at ScienceDirect

Colloids and Surfaces B: Biointerfaces

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

Zinc oxide nanoparticle and bovine serum albumin interaction and

nanoparticles inuence on cytotoxicity in vitro

Rasa Z ukien
e a,b , Valentinas Snitka a,
a
b

Research Center for Microsystems and Nanotechnology, Kaunas University of Technology, Studentu 65, Kaunas LT-51369, Lithuania
Department of Biochemistry, Vytautas Magnus University, K. Donelaicio 58, LT-44248 Kaunas, Lithuania

a r t i c l e

i n f o

Article history:
Received 29 March 2015
Received in revised form 12 July 2015
Accepted 21 July 2015
Available online 26 July 2015
Keywords:
ZnO nanoparticles
BSA
Protein structure
Raman spectroscopy
Cytotoxicity
ROS

a b s t r a c t
Bovine serum albumin (BSA) and zinc oxide nanoparticles (ZnO NPs) are chosen as a model system to
investigate NPs-protein corona complex formation. ZnO NPs with average size of 20 nm are coated
with BSA using covalent and non-covalent conjugation at temperatures of 4 C and 20 C. The interaction
mechanism between ZnO NPs and BSA is studied by using UVvis absorption, uorescence, synchronous
uorescence and Raman spectroscopy. Raman spectra of BSA in the presence of ZnO NPs are registered
for the rst time and conrm decreased -helix content, increased unstructured folding and -sheet
content in BSA structure. The synchronous uorescence spectra revealed that the hydrophobicity of the
tyrosine residue is decreased and that of the tryptophan is increased. The relation of elucidated changes
in BSA structure of BSA-coated ZnO NPs cytotoxicity is tested for CHO cell viability and reactive oxygen
species (ROS) generation in vitro. Covalent and non-covalent binding of BSA to ZnO NPs reduces ZnO NPs
cytotoxicity and ROS generation, however changes in BSA conformation makes corona less protective
against ZnO NPs.
2015 Elsevier B.V. All rights reserved.

1. Introduction
Controlling the interaction of nanomaterials with biological
interfaces is a fundamental challenge to the eld of nanomedicine,
environmental protection, health, etc. When a nanomaterial enters
a physiological environment, its surface is covered with a layer of
proteins, forming what is known as the protein corona. The protein
corona alters the size, aggregation state, and interfacial properties
of the nanomaterial, giving it a biological identity that is distinct
from its synthetic identity. Uncontrolled nanomaterialprotein
interactions can mark a nanomaterial for uptake in off-target cell
populations, activate enzymatic cascades, and prevent efcient
removal from the body [1]. A nanomaterial is safe and effective only
when its physiological response is understood and controlled. The
cosmetics or environmental inhalation exposure to the different
nanoparticles (NPs), makes a blood circulatory system to be one of
the main interaction organs exposed to NPs or NP-based nanomaterials. Therefore, a more detailed understanding of the interactions
between NPs and blood proteins may help to elucidate the potential
risk of NPs [2].

Corresponding author.
E-mail address: vsnitka@ktu.lt (V. Snitka).
http://dx.doi.org/10.1016/j.colsurfb.2015.07.054
0927-7765/ 2015 Elsevier B.V. All rights reserved.

Proteins have complicated three dimensional structures with

multi-level conformations, which are highly correlated with their
biological functions and activation of the immune response [3]. The
interaction of proteins with NPs can change their structure and
therefore disturb the proteins bio-chemical functions of various
extent. The protein-coated NPs may induce a number of biological effects, due to variation in the nature of the adsorbed or bonded
proteins or variation in the nature of the adsorbed protein conformation (in either a reversible or irreversible manner). Proteins can
offer additional functionality as targets that can be recognized by
cells and in some cases modify toxicity response proles of NPs
[4]. The concept of the nanoparticle protein corona is important
to understanding the dynamic surface properties of nanomaterials
in biological environments. In the case of protein corona formation, proteins are weakly bound to NPs by different forces (van der
Waals, hydrogen bond, electrostatic, hydrophobic, etc.), the attachment and ligand exchange reactions depend on protein properties,
NP chemical composition, size, shape, net charge and the initial
surface functionality [5]. The NPs capped with weakly bound large
proteins undergo substantial transformations due to the exchange
of the original surface ligands to the components of the body uids or cell culture media. Conversely, NP capped with covalently
conjugated proteins show signicantly higher stability. These differences in ligand exchange also affect the toxicity of the NP to the

R. Z ukien
e,

V. Snitka / Colloids and Surfaces B: Biointerfaces 135 (2015) 316323

cells, with the NP capped with small organic molecules being in

some cases more toxic than those capped with large proteins [6].
Serum albumin is the most abundant protein in mammalian
plasma and has been the subject of many investigations because
of its easy isolation in large quantities, its high stability and its
solubility. Albumin is the principal protein contributing to the colloid osmotic pressure of the blood and also the transport protein
for numerous endogenous and exogenous compounds [7]. This
molecule, having a molecular mass of 66.5 kDa, consists of a single polypeptide chain containing 583 amino acids [8]. Depending
on the analyte hydration level, solution medium and methods
used (CD, FTIR spectroscopy), the secondary BSA structure contains 34-67% -helix structure [9,10] with the remainder being
random. Some authors also report small amount (approximately
3%) of -sheet [11,12]. Serum albumins enhance cellular uptake
and internalization of NPs [13]. Nonspecic adsorption of serum
proteins mediates the uptake of the nanoparticles via nonspecic
or receptor-mediated endocytosis and dictates their fate in the
intracellular environment [14].
There are only few NPs (ZnO is one of them together with
silver and gold) known which are less toxic and biocompatible
having excellent tuned luminescence property that can be used
in several biomedical and pharmaceutical applications for human
[15,16]. ZnO is well known multifunctional wide and direct band
gap semiconductor having excellent size dependent tunable optical
property which is of great interest in the NPs based drug delivery,
bioimaging, and biomedical research [15]. However, before going
into realizing these applications, it are essential to understand the
way of interaction of ZnO NPs with blood plasma proteins, protein
corona effect on ZnO toxicity, mechanisms of toxicity and possible
ways of toxicity reduction.
In this study, bovine serum albumin (BSA) and ZnO NPs were
chosen as a model system for NP-protein corona complex formation. ZnO NPs were covalently and non-covalently coated with
BSA at 4 C and 20 C. These temperatures are usually used in
similar studies for NP coating [6,17,18]. ZnO NP-induced changes
in BSA structure and consequent effect of protein-coated NPs on
the viability and reactive oxygen species (ROS) generation in Chinese hamster ovary (CHO) cells grown in vitro were investigated.
Covalently bound BSA should not be displaced by other serum components (proteins) as in the case of non-covalently bound BSA when
used in cell culture or applied in vivo. Covalent and non-covalent
binding together with temperature induce changes in the structure
of attached protein what is an important parameter determining
the cytotoxicity of NP-protein corona complex [19]. We studied the
interaction between ZnO NPs and BSA and conformational changes
of BSA by using UV-visible absorption, uorescence, synchronous
uorescence and, for the rst time for ZnO NPs and BSA, Raman
spectroscopic measurements.

2. Materials and methods

2.1. Materials
All reagents and solvents used in this study were analytical
grade. ZnO nano-powder ZincOx 10 was obtained from Nanogate
AG (QuierschiedGttelborn, Germany), bovine serum albumin
fraction
V,
1-ethyl-3-[3-dimethylaminopropyl]carbodiimide
(EDC), N-Hydroxysuccinimide (NHS), 3-aminopropyltriethoxysilane (APTES), deionized (DI) water, 2-propanol, 3-(4,5dimethylthiazol-2-yl)-2,5-diphenyltetrazolium bromide (MTT),
2 ,7 -Dichlorodihydrouorescein diacetate (DCFH-DA), sterile
DMEM cell culture medium, fetal bovine serum (FBS), phosphate
buffered saline (PBS), l-glutamine, penicillin/streptomycin were
from SigmaAldrich (Taufkirchen, Germany).

317

2.2. Preparation of plain (ZnO), non-covalently coated (ZnOBSA)

and covalently coated (ZnOcovBSA) ZnO NP
Zinc oxide nanopowder was non-covalently and covalently
coated with BSA according to a modied procedure described by
Bartczak et al. [6]. Weighted 5 mg of powder was sonicated for
15 min at 20 C with 50 l of 2-propanol. Next, 5 ml of ultrapure
water (ZnO), 5 ml of 2% APTES in 2-propanol (ZnOcovBSA) or 5 ml
of 18,8 M BSA in DI water (ZnOBSA) was added and the suspension was sonicated for 1 h at 20 C (plain, ZnOcovBSA) or at
4 C (ZnOBSA). NPs were then reacted for additional 21 h at 20 C
(plain and ZnOBSA), 4 C (ZnOBSA) or at 60 C (ZnOcovBSA) with
shaking, prior to purication from excess ligands and/or organic
solvent residues by triple centrifugation/decantation (13,000 g,
15 min, 4 C). Capped NPs were then re-dispersed by sonication in
DI at 5 mg/ml. BSA was covalently conjugated to APTES capped NPs
using EDC/NHS coupling agents. Briey, 0.34 mg of APTES-capped
ZnO NPs were re-dispersed by sonication in 5 ml of 0.01 M borate
buffer pH 9. Then, 100 l of BSA (1 mM in DI water), 50 l of EDC
(0.198 M in DI water) and 100 l of NHS (0.198 M in DI water) were
introduced. After 21 h stirring at 20 C and 4 C, the conjugates were
puried by triple centrifugation/decantation (13,000 g, 15 min,
4 C) and re-dispersed by 15 min sonication at 4 C in DI water at
5 mg/ml.
2.2.1. Characterization of ZnO nanoparticles
Phase composition and average crystallinity of ZnO NPs was
evaluated by X-ray diffractometry (XRD). The material identication was done by the comparison of the obtained results with
the Crystallographic and Crystallochemical Database for Minerals
and their Structural Analogs (http://database.iem.ac.ru/mincryst/
index.php). The size of nanoparticles was measured by scanning
electron microscopy (SEM), atomic force microscopy (AFM) and
minimum aggregate size distribution in liquid by dynamic light
scattering (DLS).
The XRD measurements were done on Philips XPERT
diffractometer operating at 0.15406 nm wavelength (CuK1 ). A
compressed ZnO powder on the glass substrate (Hirschmann
Laborgerate) was used for the measurements.
For SEM measurement a Nova NanoSEM 630 FEG SEM (FEI, Eindhoven, the Netherlands) was used. The sample was prepared by
deposing 10 l of 1% ZnO NPs aqueous suspension on Si wafer and
dried 1 h at 60 C.
The AFM investigations were done in the tapping mode (NTMDT inc.) using commercial silicon cantilevers NSG11 with a force
constant of 5 N m1 . The ZnO water suspension (1 mg/ml) was
ltered through the 0.2 m pore diameter polyethersulfone membrane (Chromal PES-20/25, Macherey-Nagel), 25 l placed onto
the glass substrate (Carl Roth), dried and measured.
Minimum aggregate/agglomerate size distribution measurements by DLS were made using DelsaNanoC (Beckman Coulter)
tted with a size glass cell lled with 2 ml 1% ZnO suspension in
water, using detection angle of 160 .
Electrokinetic potential in colloidal systems (zeta potential)
measurements were made at 25 C on a DelsaNanoC tted with
a ow cell coupled to autotitrator. For all measurements, a eld of
60 V was applied across an electrode spacing of 16 mm, the detection angle being 15 . The solutions used for pH titration were 0.1 M
HCl and 0.1 M NaOH. The electrophoretic mobilities were converted
into zeta potential using Smoluchowskis formula [20,21].
2.3. UVvisible and uorescence studies
The UVvis absorption spectra were recorded using UVvis DU
800 (Beckman Coulter, USA) spectrophotometer. The uorescence
and synchronous uorescence spectra were recorded using LS55

318

R. Z ukien
e,

V. Snitka / Colloids and Surfaces B: Biointerfaces 135 (2015) 316323

spectrouorimeter (Perkin Elmer, USA). Excitation wavelengths

were 280 and 295 nm. For synchronous uorescence the initial
excitation wavelength was set at 200 nm and scanned up to 500 nm,
 (difference between the excitation and emission wavelengths)
was set at 15 nm or at 60 nm for tyrosine and tryptophan residues
respectively. Excitation and emission slit width was set at 5 nm in
all cases.
Fluorescence and synchronous uorescence spectra were registered in 1.88 M BSA aqueous solution, 100 g ZnO/ml ZnO NP and
BSA-coated ZnO NP aqueous suspensions, absorption spectra were
registered in two fold diluted samples.
2.4. Raman scattering spectroscopic study
Samples for Raman measurements were prepared on cleaned
glass slips by depositing 25 l of BSA (1 mM) solution or BSAcoated ZnO NPs suspension (5 mg/ml) and drying in a desiccator.
The Raman spectra are measured using a confocal Raman system
(NTEGRA Spectra, NT-MDT) with 20 mW DPSS laser, 532 nm, LCMS-111-20-NP25, a spectrometer (Solar TII, NT-MDT) equipped with
a TE-cooled (60 C) CCD camera (DV401-BV, Andor Technology,
USA) described in detail in [22]. The background correction and
noise ltering of Raman spectra were done by Raman pre-process
software [23] with following settings: median window 1, wavelet
level 1; increase in order for alternate tting 7, number of
constrained endpoints for alternate tting 5.
2.5. Curve tting of the amide I prole
Band tting of the Raman amide I mode (1610-1700 cm1 ) was
performed with the Peak Analyzer routine of OriginPro 8.5, which
is based on the Levenberg-Marquardt nonlinear least-squares
method. A Gaussian peak shape was employed. The baseline from
1610 to 1700 cm1 was assumed to be linear. Tested two-fourcomponent ts were compared by the 2 values and the residuals
of the ts were used as the criteria for assessing the quality of t.
2.6. Cell viability measurements
ZnO NP-induced cytotoxicity was evaluated by MTT assay.
CHO cells were cultured and maintained 48 h in DMEM medium
supplemented with 10% FBS, 1% l-glutamine, and 1% penicillin/streptomycin. Cells were seeded into 96-well plates with each
well receiving a volume of 200 l at a density of 104 cells/well. ZnO
NPs were sonicated for 15 min to minimize agglomeration and suspended in DMEM medium at a nal concentration of 10 g/ml. Then
100 l of each particle suspension was added to the test cells, and
100 l of supplemented medium with no ZnO NPs was added to the
control cells. Cells were cultured in plates at 37 C under a humidied atmosphere with 5% CO2 for 3 h. MTT was dissolved in DMEM at
0.5 mg/ml, and then medium with NPs was exchanged by 100 l of
this solution. The cells were incubated at 37 C for another 1 h, after
which cells were washed 3 times with PBS and 50 l of 2-propanol
was added to each well. The absorbance at 585 nm was measured
using a Tecan GeniosPro plate reader with 612 nm as a reference
wavelength. The absorbance of the background was determined
using 2-propanol but no cells. The results are presented as the percentage viability of cells exposed to ZnO NPs relative to cells not
exposed. The measurements were carried out in 16 well for each
type of NPs, experiments were repeated 3 times.
2.7. Measurements of ROS generation in cells
Levels of reactive oxygen species (ROS) generated in CHO cells
in the presence of ZnO NPs were determined by a uorometric
assay using the intracellular oxidation of DCFH-DA. Cells grown

at the same conditions as for MTT assay were incubated with

20 M DCFH-DA for 30 min and then treated with NPs (10 g/ml)
for 3 h. At the end of the incubation, cells were washed with
PBS. The uorescence of DCF (2 ,7 -dichlorouorescein), which is
the oxidized product of DCFH (2 ,7 -dichlorodihydrouorescein,
hydrolyzed from DCFH-DA by intracellular esterases), was measured with a Tecan GeniosPro plate reader using an excitation of
485 nm and an emission of 535 nm. CHO cells not exposed to NPs
were used as a control. The measurements were carried out in 16
well for each type of NPs, experiments were repeated 3 times.
2.8. Data presentation and statistical analysis
Each given spectrum is an average of three individual samples recorded twice. Data of cell viability and ROS generation are
presented as means standard error (SE) (n = 3). Statistical signicance of nanoparticles effects was evaluated using Students t-test
(paired). The differences were assumed to be statistically signicant
when p < 0.05.
3. Results and discussions
3.1. Characterization of ZnO nanoparticles
The analysis of experimental XRD spectra (not shown) peaks
were assigned to (100), (002), (101), (102), (110), (103), (200), (112),
(201), and (004) of ZnO NPs, indicating that the samples were polycrystalline wurtzite structure (Zincite, JCPDS 5-0664). SEM image
of dried drop of ZnO 1% solution in water have shown the ZnO NPs
aggregated structure. The single NPs in aggregates have an average
size about 23 nm. The grain analysis function of AFM images has
been used and then the histogram of NPs size was built. It gives
the size of ZnO of 20.2 nm. (supplementary Fig. S1). The magnitude
of the zeta potential indicates the degree of electrostatic repulsion between adjacent, similarly charged particles in dispersion.
ZnO zeta potential variation with pH 7-8.5, dispersions are stable
and positively charged (with pH 7.4 +3 mV). It is known, that low
zeta potentials tend ZnO suspension to coagulate or occulate. The
DLS measurements have shown the hydrodynamic diameter of NPs
varying from 400 nm at pH 71500 nm at pH 8.
3.2. UVvisible absorption spectroscopy
UVvisible absorption spectra of BSA solution, plain ZnO NPs
and BSA-coated ZnO NPs suspensions are shown in the inset of
Fig. 1. In sake of better differentiation and visualization of BSA
and ZnO absorption bands, the second-derivative spectra, d2 A/d2
vs. , are given in Fig. 1. BSA has a maximum absorption band at
280 nm. The absorption spectra of ZnO nanoparticles in the region
of BSA absorption depend on NP preparation (non-covalent or
covalent binding) and temperature. The difference in absorption
intensity at 280 nm may result from different amount of bound
BSA, different mechanisms of ZnO NPBSA intermolecular interactions [24], and resulting different light scattering by NPs. The 3 nm
red-shift of spectra ( d2 A/d2 ) maximum in BSA-coated ZnO NP
(283 nm) indicates different microenvironment of aromatic amino
acid residues in bound BSA as compared to free BSA (280 nm). There
was no shift of the ZnO absorption maxima at 364 nm indicating the
absence of remarkable changes in size distribution of BSA-coated
ZnO NPs and/or their aggregates as compared to plain ZnO NPs [6].
This band is not detectable when ZnO is covalently coated with
BSA (Fig. 1), most probably due to the different refractiveness [25]
of the microenvironment (i.e., protein layer) of NPs and/or scattering as compared to non-covalently coated NPs. To our knowledge,
a disappearance of this band was not previously shown; however,

R. Z ukien
e,

V. Snitka / Colloids and Surfaces B: Biointerfaces 135 (2015) 316323

319

maximum indicates the binding between ZnO NPs and BSA to form
a certain complex. To elucidate how the environment of aromatic
residues in BSA is changed from ZnO NPs binding, synchronous
uorescence analysis was done.
3.4. Synchronous uorescence of BSA in solution and on ZnO NPs

Fig. 1. Absorption spectra of BSA solution (0.94 M) and ZnO NP and BSA-coated
ZnO NP suspensions (50 g ZnO/ml).

a lower absorbtion of ZnOcovBSA as compared to ZnOBSA was

demonstrated by Bartczak et al. [6].
3.3. Fluorescence of BSA in solution and on ZnO NPs
From the uorescence spectra, we can obtain valuable information on the polarity changes around the uorophores [26]. BSA
has three intrinsic uorophores: tryptophan (Trp), tyrosine (Tyr),
and phenylalanine (Phe). Phenylalanine has a very low quantum
yield. The 295 nm light excites Trp residues, while the 280 nm light
excites both Trp and Tyr residues [27].
The uorescence emission spectra of BSA solution (free BSA)
and ZnOBSA NPs suspension at exc = 295 nm and exc = 280 nm
are shown in Fig. 2. BSA emission band maximum is at 346 nm
when exc = 280 nm. The emission maximum is blue-shifted by 10
and 11 nm in ZnOBSA NPs, 10.5 and 8.5 nm in ZnOcovBSA NPs
at 4 C and 20 C respectively. When exc = 295 nm, BSA emission
maximum is at 348 nm, the maximum is blue-shifted by 12.5 and
16 nm in ZnOBSA NPs, 13 and 11.5 nm in ZnOcovBSA NPs at 4 C
and 20 C respectively.
The main contribution to the uorescence of BSA is by Trp moiety which is very sensitive to the local environment. BSA has two
Trp residues at positions 134 and 212. Trp212 residue is located in
the largest hydrophobic cavity of the protein known as Sudlows
site I, whereas Trp134 is located on the surface of the subdomain
Ib [18]. If the binding occur to any of these domains and changes
the microenvironment (hydrophobicity) of Trp, the shift in uorescence maximum is observed. The blue-shift of the uorescence

Synchronous uorescence spectroscopy has been applied to a

variety of multi-component systems. The main advantages of synchronous uorescence spectra are simplied spectra, narrowed
bandwidth, high selectivity and sensitivity [28]. The excitation and
emission monochromators are synchronously scanned, separated
by a constant wavelength interval (). It is well known that,
synchronous uorescence spectra can provide the information on
the molecular microenvironment, particularly in the vicinity of the
uorophore functional groups [29].
The conformational changes of BSA were examined by measuring synchronous uorescence intensity of aromatic amino acid
residues in BSA and BSA-coated ZnO NPs. The environment of the
uorophore functional groups can be studied by measuring the possible shift in the maximum emission wavelength (max ). When 
is kept at 15 nm or 60 nm the synchronous uorescence spectra give
the characteristic information of Tyr or Trp residues respectively.
For this measurement, the initial excitation wavelength was set
at 200 nm and scanned up to 500 nm. Excitation and emission slit
width was set at 5 nm. Fig. 3A shows that maximum emission wavelength at  = 15 nm was red-shifted by 1.5 and 2 nm in ZnOBSA
and 3 and 2.5 nm in ZnOcovBSA at 4 C and 20 C respectively. The
maximum emission wavelength at  = 60 nm was blue-shifted
by 3 and 2 nm in ZnOBSA and 3 and 2.5 nm in ZnOcovBSA at
4 C and 20 C respectively (Fig. 3B). This result demonstrates that
the microenvironment close to both Tyr and Trp residues of BSA is
perturbed, the hydrophobicity of Tyr residue is decreased and that
of Trp residue is increased in the presence of ZnO NPs. Covalent
binding of BSA to ZnO NPs creates less hydrophobic Tyr environment as compared to non-covalent binding at both temperatures
used. NPs-induced increase in Trp hydrophobicity was similar at
4 C for non-covalent and covalent binding and was lower at 20 C,
especially for non-covalently-bound BSA.
Bhogale et al. showed the hydrophobicity of both Tyr and Trp
residues were increased in the presence of ZnO NPs [30]. This
discrepancy with our data of Tyr may result from different concentrations of BSA, ZnO NPs size and NP incubation time with
BSA used. Study of the binding of colloidal ZnO NPs to BSA by
Kathiravan et al. [28] showed that the emission wavelength of
Tyr residues was blue-shifted with increasing concentration of
colloidal ZnO nanoparticles meaning the conformation of BSA is
somewhat changed, leading to the polarity around Tyr residues
strengthened and the hydrophobicity weakened [31]. At the same

Fig. 2. Fluorescence spectra of BSA solution (1.88 M), ZnO NP and BSA-coated ZnO NP suspensions (100 g ZnO/ml). A c = 295 nm, B exc = 280 nm.

320

R. Z ukien
e,

V. Snitka / Colloids and Surfaces B: Biointerfaces 135 (2015) 316323

Fig. 3. Synchronous uorescence spectra of BSA solution (1.88 M), ZnO NP and BSA-coated ZnO NP suspensions (100 g ZnO/ml) at: A  = 15 nm, B  = 60 nm.

time, the Trp uorescence decreased regularly, but no signicant

change in a wavelength was observed. Authors suggest that the
interaction of colloidal ZnO nanoparticles with BSA does not affect
the conformation of Trp micro-region. The proposed mechanism by
which Tyr residue interacts with ZnO NPs is through the aromatic
hydroxyl group of Tyr which is absent in Trp [28].

3.5. Raman spectroscopy

The measured Raman spectra of BSA, ZnOBSA, and ZnOcovBSA
NPs, prepared at 4 C and 20 C, are given in supplementary Fig.
S2. Visually detectable differences between Raman spectra components of NPs prepared at different temperatures can be observed in
the region of 870-1000 cm1 already from measured spectra. The
same spectra after pre-processing (baseline correction, cosmic rays
and noise ltering) are shown in Fig. 4A and B, respectively.
The amide I band is a carbonyl stretching mode vibration, and
amide III band combines both in-plane N H bending and C N
stretching motions are the most useful spectra components for
the protein secondary structure analysis by Raman spectroscopy.
Amide I and amide III bands are expected in the spectral intervals of 16401680 cm1 and 12301310 cm1 , respectively, and
the band shapes and peak positions are sensitive to the protein
secondary structure. Accordingly, these Raman amide bands are
often used as indicators of protein secondary structure [32]. The
amide I deconvoluted spectral region is shown in supplementary
Fig. S3 and corresponding structure content is given in Table 1. In
our case, amide I band of 1654-1657 cm1 corresponding to an helix structure is the most intensive in free BSA, the content of
-helix is 13.1% and 2.9% lower at 4 C and 9.2% and 3.4% lower
at 20 C in non-covalently and covalently bound BSA, respectively
(Table 1). This indicates that -helix content decreases more in the
case of non-covalent as compared to covalent BSA binding at both
temperatures applied.
The band maxima of -helix in amide III region (1300 and
1272 cm1 in free BSA) is shifted to 1292 and 1267 cm1 , respectively in the case of BSA binding to ZnO NPs. The lower frequency
regions represent less compact structures (lower -helix content).
The shift in the regions of coils, turns and -sheets is characteristic
for both, non-covalently and covalently bound BSA, only the band
maximum is shifted more (by 2-3 cm1 ) in the case of ZnOBSA
as compared to ZnOcovBSA, indicating more unstructured folding
and more -sheet structures. These results are in a good agreement
with amide I region ndings conrming the conformation of BSA
on ZnO NPs is changed (less -helix) and BSA conformation being
more inuenced by ZnO NPs in the case of non-covalent binding.

Fig. 4. The pre-processed Raman spectra of BSA, ZnOBSA, and ZnOcovBSA

nanoparticles prepared at: A 4 C, B 20 C.

This could be possibly explained by less contact of BSA with ZnO

NPs in covalent binding since NPs are coated with APTES.
Other important information which can be derived from Raman
spectra beside amide group vibrations is the vibrations of aromatic
amino acid (Tyr, Trp, Phe) residues. It has long been recognized
that the para-substituted phenyl ring of tyrosine can generate a

R. Z ukien
e,

V. Snitka / Colloids and Surfaces B: Biointerfaces 135 (2015) 316323

321

Table 1
Assignments of bands in amide I region of BSA, ZnOBSA, and ZnOcovBSA NPs spectra.
Temperature

4 C

20 C

Sample

Structure
Raman shift (t peak area (%))
-Helix

Random coil

-Strand

Extended conformation

BSA
ZnOBSA
ZnOcovBSA

1654 (83.2)
1654 (70.1)
1656 (80.4)

1665 (4.1)

1672 (10.7)
1676 (9.3)

1679 (16.8)
1683 (15.1)
1686 (10.3)

BSA
ZnOBSA
ZnOcovBSA

1654 (83.0)
1655 (74.0)
1657 (79.8)

1670 (5.9)
1669 (1.4)

1677 (8.4)
1677 (4.8)

1679 (17.0)
1687 (11.7)
1687 (14.0)

pair of relatively intense Raman bands at 850 and 830 cm1 [33].
Siamwiza et al. demonstrated [34] that the ratio I850 /I830 reects
the OH hydrogen-bonding state. If the protein contains multiple tyrosines, as BSA (20), I850 /I830 reects the average of tyrosyl
hydrogen-bonding states [31]. In our case, the I850 /I830 ratio for BSA
is 1.26, for ZnOBSA 1.21 and for ZnOcovBSA 1.08. It means
that the phenolic OH group of free BSA functions as both a donor
and an acceptor whereas the phenolic OH group of bound BSA is
more a hydrogen-bond donor, being a stronger donor in covalently
bound BSA. The synchronous uorescence results also conrm the
changes in tyrosine microenvironment due to the conformational
changes of BSA and/or interaction with ZnO NPs: the hydrophobicity of tyrosine residues is decreased.
The indolyl moiety of Trp generates many prominent Raman
bands [35], several of which have been correlated with the local
environment and geometry of Trp side chain in proteins [30]. In
our case, the maximum of Trp band of both free and bound BSA
is 1553 cm1 , indicating the mean value of the side chain torsion
angle of 2 Trp residues being unchanged in the interaction with ZnO
NP.
Our results of synchronous uorescence measurements demonstrate that the microenvironment close to the Trp residues is
perturbed and that the hydrophobicity of Trp residue is increased
from the presence of ZnO NPs. These results are in agreement with
other authors results [29] and are proved by our Raman spectra: Trp residue generates a Fermi doublet with components at
1360 and 1340 cm1 . The Fermi doublet intensity ratio I1360 /I134
increases with increasing hydrophobicity of the indolyl ring environment and thus serve as an indicator of local hydropaty [36]. The
ratio I1360 /I1340 is 0.10 for BSA, 0.22 for ZnOBSA and 0.17 for
ZnOcovBSA, thus, conrming the results, obtained by synchronous
uorescence analysis: ZnO NP increase the hydrophobicity of Trp
residues.
The band of Phe in free BSA Raman spectrum is at 1002 cm1 ,
in ZnOBSA and ZnOcovBSA spectra at 1005 cm1 , indicating
different environment of Phe residue. This was partly conrmed
by uorescence spectra (ex = 280 nm, based on all aromatic amino
acids residues in BSA) where the uorescence maximum (346 nm
in free BSA) was blue-shifted by 10 nm in the presence of ZnO NP.
The stretching vibrations of COO generate a signal at
1406 cm1 . The band is less prominent in ZnOBSA than in free
BSA (intensities are 0.18 and 0.26, respectively) and disappears in
ZnOcovBSA spectrum. This may indicate the decreased amount
of free side COO groups of aspartic and glutamic acid residues
due to their involvement in covalent bond formation with APTEScoated ZnO NPs.
Raman spectra of BSA-coated NPs, prepared at 20 C, have additional changes as compared to 4 C: there is a prominent band
at 731 cm1 of C S group asymmetric stretching and additional
band at 702 cm1 of C S group stretching in ZnOcovBSA spectrum, meaning disulphide bridge breakage, decrease in 940 cm1
band assigned to -helix, decrease in 898, 960 and 1446 cm1 bands
assigned to lysine and leucine residues vibrations [34].

Fig. 5. Effect of 3 h incubation with ZnO, ZnOBSA and ZnOcovBSA NPs (10 g
ZnO/ml) on CHO cells: A viability determined by MTT assay, B ROS generation determined by DCF uorescence. Mean SE (n = 3). * statistically signicant
difference compared to control (p < 0.05).

3.6. ZnO NP-induced cytotoxicity in Chinese hamster ovary (CHO)

cells
The effects of 3 h exposure to plain and BSA-coated ZnO NPs
(10 g/ml) on CHO cell viability is shown in Fig. 5A, where the viability of control cells (no NP treatment) is normalized to 100%. Cell
viability was assessed by the MTT assay, which measures mitochondrial reductase activity. Cell viability in response to treatment with
plain ZnO NPs differs from that observed in response to the BSAcoated NPs. Cellular activity was reduced by 11% in the presence
of plain ZnO NPs. Non-covalently and covalently BSA-coated ZnO
NPs were non-toxic to CHO cells when prepared at 4 C. BSA-coated
ZnO NPs, prepared at 20 C were slightly cytotoxic: ZnOBSA and
ZnOcovBSA decreased viability by 6% and 7%, respectively. However, the cytotoxicity was statistically signicantly lower than that
of plain ZnO NPs. Our results do not agree with those obtained by
Barczak et al. showing strongly reduced cytotoxicity by covalent
BSA binding to ZnO NPs as compared to non-covalent binding in
HepG2 liver cell model using impedance spectroscopy monitoring
[5]. Our results conrm the ndings of Yin et al., were ZnO NPs
coated with oleic acid, polymethacrylic acid, or components from
the supplemented RPMI 1640 cell grow medium were tested on
WIL2-NS human lymphoblastoid cells and had lower cytotoxicity
than plain ZnO NPs [16].

322

R. Z ukien
e,

V. Snitka / Colloids and Surfaces B: Biointerfaces 135 (2015) 316323

There is still very little information to which extent protein

coating (protein corona of one or multiple proteins) decreases or
diminishes toxicity of plain ZnO NPs at relatively low concentration at real exposure and what is the effect of spontaneous binding
and covalent immobilization. More are known about the mechanisms of plain ZnO NPs actions: ZnO induce decrease in cell viability
by inducing apoptosis, polymerization of F-actin, ROS generation,
reduced glutathione (GSH) depletions, and morphological changes
in nucleus [36]. It was shown that Zn+ ions at the maximum dissolution concentrations do not cause cytotoxicity, ZnO NPs do not
disturb membrane integrity but activates multi-drug resistance
pump MRP-1/PgP which export glutathione from the cell and can
be a reason of apoptosis [37].
3.7. ZnO NP-induced ROS generation in CHO cells
It has been shown that oxidative stress plays a key role in
the development of toxicity in response to nanoparticle challenge
[3639] and, in most cases, protein corona decreases ROS generation, characteristic to plain NPs [40]. Levels of ROS, generated by
ZnO NPs in CHO cells are shown in Fig. 5B.
ZnOBSA and ZnOcovBSA NPs, prepared at 4 C, do not induce
ROS generation. ROS generation was induced by plain ZnO NPs and
by ZnOBSA and ZnOcovBSA NPs, prepared at 20 C by 13%, 6% and
5%, respectively. Increased ROS generation correlates with slightly
decreased viability (Fig. 5A).
BSA-coated ZnO NPs, prepared at 20 C, were more cytotoxic
and induced ROS generation what was not a case for NPs, prepared
at 4 C. The reason could be additional ZnO NPs-induced changes
in BSA structure at 20 C as was dened by Raman spectroscopy.
In the case of non-covalent bonding these changes may result in
lower BSA afnity to ZnO NPs what can make BSA easier replaceable
by cell medium components or intracellular components. When
BSA is covalently bound to NPs, 20 C temperature-induced protein
structural changes can be recognized by extracellular or intracellular receptors inducing ROS generation and consecutive apoptosis.
However, the exact mechanism should be further investigated.
4. Conclusions
ZnO NPs-induced conformational changes in BSA manifest in
a decreased -helix content, increased part of unstructured folding and -sheet structures, and in a changed microenvironment
of aromatic amino acid (Tyr, Trp, Phe) residues. Covalent binding of BSA to ZnO NPs induces similar changes in BSA structure
and forms the protein corona which has the same biological effect
(reduce ZnO NPs-induced cytotoxicity and ROS generation) as noncovalently formed. More crucial parameter than binding mode for
NPs cytotoxicity is the NPs-coating temperature: 20 C temperature
induces additional changes in BSA (disulphide bridge breakage)
which make corona less protective against ZnO NPs as compared
to 4 C.
Acknowledgements
This research is funded by the European Social Fund under the
Global Grant measure (Grant No. VP1-3.1-SMM-07-K-03-044).
Appendix A. Supplementary data
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.colsurfb.2015.07.
054

References
[1] P. del Pino, B. Pelaz, Q. Zhang, P. Maffre, G.U. Nienhaus, W.J. Parak, Protein
corona formation around nanoparticles from the past to the future, Mater.
Horiz. 1 (2014) 301313.
[2] S. Halappanavar, A.T. Saber, N. Decan, K.A. Jensen, D. Wu, N.R. Jacobsen, C.
Guo, J. Rogowski, I.K. Koponen, M. Levin, A.M. Madsen, R. Atluri, V. Snitka, R.K.
Birkedal, D. Rickerby, A. Williams, H. Wallin, C.L. Yauk, U. Vogel,
Transcriptional proling identies physicochemical properties of
nanomaterials that are determinants of the in vivo pulmonary response,
Environ. Mol. Mutagen (2014), http://dx.doi.org/10.1002/em.21936
[3] I. Lynch, K.A. Dawson, S. Linse, Detecting cryptic epitopes created by
nanoparticles, Sci. STKE 327 (2006) pe14.
[4] T. Xia, M. Kovochich, M. Liong, L. Madler, B. Gilbert, H. Shi, J.I. Yeh, J.I. Zink, A.E.
Nel, Comparison of the mechanism of toxicity of zinc oxide and cerium oxide
nanoparticles based on dissolution and oxidative stress properties, ACS Nano
2 (10) (2008) 21212134.
[5] C. You, A. Chompoosor, V.M. Rotello, The biomacromolecule-nanoparticle
interface, Nanotoday 2 (3) (2007) 3443.
[6] D. Bartczak, M.-O. Baradez, S. Merson, H. Goenaga-Infante, D. Marshall,
Surface ligand dependent toxicity of zinc oxide nanoparticles in HepG2 cell
model, J. Phys.: Conf. Ser. 429 (2013) 012015.
[7] T. Peters Jr., All About Albumin, Academic Press, New York, 1996, pp. 975.
[8] G. Das, F. Gentile, M.L. Coluccio, A.M. Perri, A. Nicastri, F. Mecarini, G. Cojoc, P.
Candeloro, C. Liberale, F. De Angelis, E. Di Fabrizio, Principal component
analysis based methodology to distinguish protein SERS spectra, J. Mol. Struct.
993 (13) (2011) 500505.
[9] B. Bolton, J.R. Scherer, Raman spectra and water absorption of bovine serum
albumin, J. Phys. Chem. 93 (1989) 76357640.
[10] D.C. Carter, J.X. Ho, Structure of serum albumin, Adv. Protein Chem. 45 (1994)
153203.
[11] S. Saha, T. Kamilya, R. Bhattacharya, A. Bhunia, Unfolding of blood plasma
albumin protein in interaction with CdS nanoparticles, Sci. Adv. Mater. 6 (1)
(2014) 5662.
[12] T. Kamilya, P. Pal, M. Mahato, G.B. Talapatra, Fabrication of
ovalbuminphospolipid thin lm with minimal protein aggregation by
different self-assembly methods, J. Nanosci. Nanotechnol. 9 (5) (2009)
29562964.
[13] S.H. Kim, J.H. Jeong, K.W. Chun, G. Park, Target-specic cellular uptake of
PLGA nanoparticles coated with poly(L-lysine)-poly(ethylene glycol)-folate
conjugate, Langmuir 21 (2005) 88528857.
[14] B.D. Chithrani, A.A. Ghazani, W.C.W. Chan, Determining the size and shape
dependence of gold nanoparticle uptake into mammalian cells, Nano Lett. 6
(2006) 662668.
[15] J.W. Rasmussen, E. Martinez, P. Louka, D.G. Wingett, Zinc oxide nanoparticles
for selective destruction of tumor cells and potential for drug delivery
applications, Expert. Opin. Drug Deliv. 7 (9) (2010) 10631077.
[16] P.K. Brown, A.T. Qureshi, A.N. Moll, D.J. Hayes, W.T. Monroe, Silver nanoscale
antisense drug delivery system for photoactivated gene silencing, ACS Nano 7
(4) (2013) 29482959.
[17] H. Yin, P.S. Casey, M.J. McCall, M. Fenech, Effects of surface chemistry on
cytotoxicity, genotoxicity, and the generation of reactive oxygen species
induced by ZnO nanoparticles, Langmuir 26 (19) (2010)
1539915408.
[18] S. Chakraborti, P. Joshi, D. Chakravarty, V. Shanker, Z.A. Ansari, S.P. Singh, P.
Chakrabarti, Interaction of polyethyleneimine-functionalized ZnO
nanoparticles with bovine serum albumin, Langmuir 28 (2012) 1114211152.
[19] S.R. Saptarshi, A. Duschl, A.L. Lopata, Interaction of nanoparticles with
proteins: relation to bio-reactivity of the nanoparticle, J. Nanotechnol. 11
(2013) 26.
[20] M. Smoluchowski, Handbuch der Electrizitt und des Magnetismus (Graetz),
vol. II, Barth, Leipzig, 1921, pp. 366.
[21] A.V. Delgado, F. Gonzlez-Caballero, R.J. Hunter, L.K. Koopal, J. Lyklema,
Measurement and interpretation of electrokinetic phenomena, Pure Appl.
Chem. 77 (2005) 17531805.
[22] D. Naumenko, V. Snitka, E. Serviene, I. Bruzaite, B. Snopok, In vivo
characterization of protein uptake by yeast cell envelope: single cell AFM
imaging and -tip-enhanced Raman scattering study, Analyst 138 (18) (2013)
53715383.
[23] L.A. Reisner, A. Cao, A.K. Pandya, An integrated software system for
processing, analyzing, and classifying Raman spectra, Chemom. Intell. Lab.
Syst. 105 (2011) 8390.
[24] M. Bardhan, G. Manda, T. Ganguly, Steady state, time resolved, and circular
dichroism spectroscopic studies to reveal the nature of interactions of zinc
oxide nanoparticles with transport protein bovine serum albumin and to
monitor the possible protein conformational changes, J. Appl. Phys. 106
(2009) 034701.
[25] K.L. Kelly, E. Coronado, L.L. Zhao, G.C. Schatz, The optical properties of metal
nanoparticles: the inuence of size, shape and dielectric environment, J. Phys.
Chem. B 107 (2003) 668677.
[26] M. Iosin, F. Toderas, P.L. Baldeck, S. Astilean, Study of protein-gold
nanoparticle conjugates by uorescence and surface-enhanced Raman
scattering, J. Mol. Struct. 924926 (2009) 196200.
[27] A. Sulkowska, Interaction of drugs with bovine and human serum albumin, J.
Mol. Struct. 614 (2002) 227232.

R. Z ukien
e,

V. Snitka / Colloids and Surfaces B: Biointerfaces 135 (2015) 316323

[28] A. Kathiravan, G. Paramaguru, R. Renganathan, Study on the binding of
colloidal zinc oxide nanoparticles with bovine serum albumin, J. Mol. Struct.
(2009) 129137.
[29] G.Z. Chen, X.Z. Huang, J.G. Xu, Z.B. Wang, Z.Z. Zhang, Method of Fluorescent
Analysis, 2nd ed., Science Press, Beijing, 1990, pp. 123126.
[30] A. Bhogale, N. Patel, P. Sarpotdar, J. Mariam, P.M. Dongre, A. Miotello, D.C.
Kathari, Systematic investigation on the interaction of bovine serum albumin
with ZnO nanoparticles using uorescence spectroscopy, Colloids Surf. B 102
(2013) 257264.
[31] B. Klajnert, M. Bryszewska, Fluorescence studies on PAMAM dendrimers
interactions with bovine serum albumin, Bioelectrochemistry 55 (12) (2002)
3335.
[32] D. Nemecek, J. Stepanek, G.J. Thomas Jr., Spectroscopy of proteins and
nucleoproteins, Curr. Protoc. Protein Sci. 71 (2013) 17.8.117.8.52.
[33] R.C. Lord, N.T. Yu, Laser-excited Raman spectroscopy of biomolecules. Native
lysozyme and its constituent amino acids, J. Mol. Biol. 50 (1970) 509524.
[34] M.N. Siamwiza, R.C. Lord, M.C. Chen, T. Takamatsu, I. Harada, H. Matsuura, T.
Shimanouchi, Interpretation of the doublet at 850 and 830 cm1 in the Raman

[35]

[36]
[37]

[38]
[39]
[40]

323

spectra of tyrosyl residues in proteins and certain model compounds,

Biochemistry 14 (1975) 48704876.
B. Sjberg, S. Foley, B. Cardey, M. Enescu, An experimental and theoretical
study of the amino acid side chain Raman bands in proteins, Spectrochim.
Acta A: Mol. Biomol. Spectrosc. 128 (2014) 300311.
I. Harada, H. Takeuchi, Raman and ultraviolet resonance Raman of proteins
and related compounds, Adv. Spectrosc. 13 (1986) 113175.
J.M. Seiffert, M.-O. Baradez, V. Nischwitz, T. Lekishvili, H. Goenaga-Infante, D.
Marshall, Dynamic monitoring of metal oxide nanoparticle toxicity by label
free impedance sensing, Chem. Res. Toxicol. 25 (2012) 140152.
A. Nel, T. Xia, L. Mdler, Toxic potential of materials at the nanolevel, Science
311 (5761) (2006) 622627.
V. Stone, K. Donaldson, Nanotoxicology: signs of stress, Nat. Nanotechnol. 1
(2006) 2324.
N.P. Sasidharan, P. Chandran, S.S. Khan, Interaction of colloidal zinc oxide
nanoparticles with bovine serum albumin and its adsorption isotherms and
kinetics, Colloids Surf. B 102 (2013) 195201.