The bilateral action between EQ14-2-14 gemini surfactant and bovine serum albumin by DPI and 1H NMR

Article abstract of DOI:10.1016/j.molstruc.2013.04.003

Gemini surfactant diglycol bis-N-tetradecyl nicotinate dibromide (designed as EQ14-2-14) has been synthesized. The interaction between EQ14-2-14 and bovine serum albumin (BSA) was studied by dual polarization interferometry (DPI), proton nuclear magnetic resonance spectroscopy (¹H NMR), Fourier transform infrared spectroscopy (FTIR) and molecular docking. Owing to the binding of EQ14-2-14, the thickness and mass of BSA increased; refractive index (RI) and density firstly raised and then tended to a plateau. In addition, a decrease of α-helix was observed from 54.01% to 31.56% with an increase in random structure from 7.86% to 21.76%. Due to BSA intertwining, the proton resonance signals of EQ14-2-14 shifted up-field and relaxation time decreased with increasing concentration of BSA. The study of molecular docking indicated that EQ14-2-14 embedded into subdomain II of BSA by π-π stacking between the electron-deficit pyridinium rings in EQ14-2-14 and the electron-abundant pyrrole ring in Trp residues of BSA, by hydrogen bonding and by hydrophobic interaction. Therefore the present work offers a whole view of the interaction of BSA with a new gemini surfactant.

Full text of DOI:10.1016/j.molstruc.2013.04.003

Journal of Molecular Structure 1045 (2013) 47–54
Contents lists available at SciVerse ScienceDirect
Journal of Molecular Structure
journal homepage: www.elsevier.com/locate/molstruc
The bilateral action between EQ14-2-14 gemini surfactant and bovine
1
serum albumin by DPI and H NMR
a
Gang Wu a, Xiaohui Jiang a, , Limei Zhou , Lijun Yang a,c, Ya Wang a, Guangqiang Xia a, Zhengjun Chen a,
⇑
Ming Duan b
a Chemical Synthesis and Pollution Control Key Laboratory of Sichuan Province, China West Normal University, Nanchong, Sichuan 637009, PR China
b State Key Laboratory of Oil and Gas Reservoir Geology and Exploitation, Southwest Petroleum University, Chengdu, Sichuan 610500, PR China
c State Key Laboratory of Biotherapy, West China Hospital, Sichuan University, Chengdu 610041, PR China
h i g h l i g h t s
ꢀ
ꢀ
ꢀ
ꢀ
EQ14-2-14 is a newly synthesized cationic gemini surfactant.
Combining with EQ14-2-14 changes the thickness, refractive index, density and mass of BSA layer.
Binding by BSA causes 1H signals of EQ14-2-14 up-field shifted and relaxation time decreased.
p–p stacking, hydrogen bonding and hydrophobic interaction are responsible for the binding.
a r t i c l e i n f o
a b s t r a c t
Article history:
Gemini surfactant diglycol bis-N-tetradecyl nicotinate dibromide (designed as EQ14-2-14) has been syn-
thesized. The interaction between EQ14-2-14 and bovine serum albumin (BSA) was studied by dual
polarization interferometry (DPI), proton nuclear magnetic resonance spectroscopy (1H NMR), Fourier
transform infrared spectroscopy (FTIR) and molecular docking. Owing to the binding of EQ14-2-14, the
thickness and mass of BSA increased; refractive index (RI) and density firstly raised and then tended
Received 29 January 2013
Received in revised form 31 March 2013
Accepted 1 April 2013
Available online 8 April 2013
to a plateau. In addition, a decrease of
in random structure from 7.86% to 21.76%. Due to BSA intertwining, the proton resonance signals of
EQ14-2-14 shifted up-field and relaxation time decreased with increasing concentration of BSA. The
study of molecular docking indicated that EQ14-2-14 embedded into subdomain II of BSA by p–p stack-
ing between the electron-deficit pyridinium rings in EQ14-2-14 and the electron-abundant pyrrole ring
in Trp residues of BSA, by hydrogen bonding and by hydrophobic interaction. Therefore the present work
offers a whole view of the interaction of BSA with a new gemini surfactant.
Ó 2013 Elsevier B.V. All rights reserved.
a-helix was observed from 54.01% to 31.56% with an increase
Keywords:
Cationic gemini surfactant
Bovine serum albumin
Proton nuclear magnetic resonance
spectroscopy
Dual polarization interferometry
Molecular docking
1. Introduction
and in the mixture of BSA-surfactant, by which they could judge
if the secondary structure of BSA is altered by surfactant binding
Protein-surfactant interaction has been a subject of intensive
studies for decades, because surfactants are widely used in indus-
try and agriculture process and in daily life [1–6]. Because bovine
serum albumin (BSA) and human serum albumin (HSA) have a se-
quence homology of 76% and have similar 3D structures [7], BSA is
an appropriate protein model to study the interaction of protein
and ligands.
[8,9]. According to the change in the intensity and emission posi-
tion of fluorescence spectra, many researchers have reported the
binding constants Ka of BSA with surfactants [10,11], the thermo-
dynamic parameters such as
D D D
G0, H0 and S0, the binding site I
or site II, and the binding site number n [12–14]. Overlapping spec-
tra of fluorescence and UV–vis offers the distance between the do-
nor (BSA) and the acceptor (surfactants) as well as the energy
transfer efficiency [15–17]. In addition, the fluorescence quenching
mechanism, either static or dynamic quenching, can be assessed by
measuring the fluorescence lifetime [18]. However, these methods
reveal some changes of protein microstructure, the other
macroproperties, such as the density and thickness of BSA, cannot
be obtained, which are certainly changed due to the combination
with surfactants.
By means of FTIR and circular dichroism (CD), people get the
information of the changed secondary structure of BSA, and the
contents of
a-helix, b-turning and sheets as well as the random
structure can be quantitatively calculated both in BSA solution
⇑
Corresponding author. Tel.: +86 817 2314508; fax: +86 817 2568625.
E-mail address: jxh2314508@163.com (X. Jiang).
0022-2860/$ - see front matter Ó 2013 Elsevier B.V. All rights reserved.
http://dx.doi.org/10.1016/j.molstruc.2013.04.003

48
G. Wu et al. / Journal of Molecular Structure 1045 (2013) 47–54
While the structure of BSA varies upon the addition of surfac-
Mix diethylene glycol diester (0.015 mol) and bromotetradec-
ane (0.045 mol) with acetonitrile, and reflux the solution for
40 h. Recrystallize the product in dichloromethane/acetone (1:20)
at least three times, collect white powder (the gemini surfactant)
and dry it in vacuum at 30 °C for 12 h. The structure of EQ14-2-
14 was confirmed by MS spectrometry and characterized by melt-
ing point, IR and NMR spectrometry. (Figs. S1–S4).
tants, there surely are changes happening inside the surfactants.
Zhou et al. [19] studied the complexes of C.I. Acid Orange 7 with
HSA by 1H NMR. All the chemical shifts of 1H signals moved
up-field and the spin–lattice relaxation times (T1) of C.I. Acid Or-
ange 7 decreased by the addition of HSA. The data confirmed that
C.I. Acid Orange 7 indeed interacted with HSA, and hydrophobic
portion of C.I. Acid Orange 7 should be embedded into the hydro-
phobic pocket of HSA. Research by Wu et al. [20] indicated that all
the 1H signals of naphthol shifted up-field when adding BSA, which
EQ14-2-14: white powder; Yield: 62%; MP (°C): 77–78. 1H NMR
(400 MHz, D2O, TMS): d 9.614(s, 1H), 9.360 (d, 1H, J = 4.4 Hz), 9.107
(d, 1H, J = 8 Hz), 8.357–8.326 (dd, 1H, J1 = 6.4 Hz, J2 = 6 Hz), 4.883
(s, 2H), 4.589 (s, 2H), 4.008 (s, 2H), 2.004 (s, 2H), 1.257–1.105
indicated p–p stacking occurring between naphthol and aromatic
residues of BSA. And the protons far from the hydroxyl group
up-field shifted significantly, which suggested that these protons
were close to the aromatic ring in the binding site. However, little
study is reported referring to BSA and gemini surfactants owing to
the expensive instrument, dear reagents, complicated method and
so on.
(m, 22H), 0.627 (s, 3H). 13C NMR (100 MHz, CDCl3,TMS):
d
161.57, 148.38, 145.78, 145.59, 130.34, 129.06, 68.71, 66.04,
62.55, 31.89, 31.85, 29.68, 29.66, 29.63, 29.58, 29.42, 29.34,
29.17, 26.09, 22.66, 14.09; FT-IR (KBr, cmꢁ1): 3425, 2924, 2854,
1736, 1638, 1462, 1378, 1301, 1178,1122, 1023,872, 749, 680,
539. MS (m/z): [M-Br]+ 355.2798.
Therefore, the purpose of the present work is to synthesize a
new gemini surfactant (EQ14-2-14) and probe the bilateral action
between BSA and the gemini surfactant by DPI, FTIR, 1H NMR and
molecular docking, so as to get a more comprehensive understand
about the interaction between proteins and gemini surfactants.
2.3. FTIR spectroscopy
All FTIR spectra were taken on a Nicolet 6700 FTIR spectrometer
(ThermoFisher Scientific, Massachusetts, USA) via the attenuated
total reflection (ATR) with resolution of 4 cmꢁ1 and 64 scans. The
spectrum of D2O was collected first, and then it was subtracted
from the spectrum of EQ14-2-14/D2O solution to get the FTIR spec-
trum of EQ14-2-14. The spectrum of EQ14-2-14/D2O was sub-
tracted from that of BSA/EQ14-2-14/D2O to get the FTIR
spectrum of BSA binding with EQ14-2-14. A quantitative analysis
of the spectra of BSA before and after the addition of EQ14-2-14
was processed by Fourier self-deconvolution, second-derivation
and Gaussian curve-fitting.
2. Materials and methods
2.1. Materials
Nicotinic acid (CR), thionyl chloride (AR), diglycol (AR) and bro-
motetradecane (AR) were purchased from Chengdu Kelong Chem-
ical Reagent Company, BSA from Shanghai Fankel Biological
Technology CO., Ltd., tris–HCl buffer (1.0 M, pH 7.4) from Beijing
Solarbio Science & Technology CO., Ltd., and deuterium oxide with
purity of 99.994% from Qingdao Tenglong Weibo Technology Co.,
Ltd. All other chemicals were of analytical grade and used as sup-
plied without further purification.
2.4. Nuclear magnetic resonance measurements
All solutions of EQ14-2-14 and BSA were prepared at pH 7.4 in
0.05 M sodium phosphate buffer in D2O. The concentration of
EQ14-2-14 stock solution was 1.0 ꢂ 10ꢁ2 M. The concentration of
BSA was 5 ꢂ 10ꢁ3 M. 1H NMR measurements at different molar ra-
tio ([BSA]:[EQ14-2-14]) were recorded on a Bruker Advance 400
spectrometer (Rheinstetten, Germany) by using a probe tuned at
400 MHz and 5 mm tubes at 298 K. Chemical shift of EQ14-2-14
proton resonances were measured in relation to the TMS signal
at 0 ppm. Presaturation method was used for water suppression.
T1 of protons were determined by conventional inversion-recovery
method. Maximum error of chemical shift and relaxation time was
0.0001 ppm and 0.001 s, respectively.
2.2. Preparation and characterization of EQ14-2-14
Scheme 1 presents the way for preparing the gemini surfactant.
The procedure is as follow.
Add thionyl chloride (0.2 mol) to nicotinic acid (0.1 mol) under
stirring, and reflux the mixture for 4 h. Remove excess thionyl
chloride by a rotary evaporator, add dropwise the mixture of digly-
col (0.045 mol) and chloroform into the flask under agitating at
65 °C, reflux the mixture for 10 h, then evaporate the solvent.
Add saturated Na2CO3 solution to the residue to adjust the solution
to neutrality or alkalescence. Extract the mixture with ethyl ace-
tate, then, purify the extractive by column chromatography (petro-
leum ether: ethyl acetate = 3:1) and collect pale yellow liquid
(Diethylene glycol diester).
2.5. DPI measurements
The details of Dual polarization interferometry (Ana Light
Nano2000, Farfield Group Ltd., Manchester, U.K.) and theory has
Scheme 1. Preparation of the EQ14-2-14.

G. Wu et al. / Journal of Molecular Structure 1045 (2013) 47–54
49
been described previously by Karim et al. (Fig. S5) [21]. Each sensor
chip’s active surface is unmodified siliconoxynitride. A PHD 2000
Infusion pump (Harvard Apparatus) was used to provide liquid
injection. The experiment was started by the injection of BSA solu-
JYW-200C tensiometer (Chengde Testing Machine Co., Ltd., China)
using the ring method, respectively. The average values of equilib-
rium surface tension were obtained by repeating three times. Max-
imum error of surface tension was 0.1 mN/m.
tion (20 lM, pH 7.4). The solution of EQ14-2-14 (63.24 lM) was
then injected to the chip surface followed by rinsing with phos-
phate buffer until the signal became stable. All solutions used in
the DPI experiments were degassed to get rid of air bubbles in
the flow system, and the temperature was maintained at 293 K
throughout the measurement.
The obtained data were analyzed in AnaLight Explorer (Farfield
Scientific) to obtain thickness, density and mass values. The DPI
technique is based on the measurement of two polarized beam sig-
nals the transverse electric (TE) and the transverse magnetic (TM),
determined from the vertical and horizontal directions, respec-
tively. By solving the Maxwell’s equations for both responses as a
function of time, both the thickness and refractive index (RI) can
be determined. RI provides information on the density of the depo-
sition layer; thus, the mass of the adsorbed layer can be calculated
from the following equation:
3. Results and discussion
3.1. The change of BSA conformation caused by EQ14-2-14
3.1.1. DPI analysis
TE&TM plots of BSA and EQ14-2-14 at the hydrophilic silicon
oxide/solution interface are depicted in Fig. S6. Fig. 1 graphs the
adsorption properties of BSA (20
immobilization of EQ14-2-14 (63.24
l
M) on the sensor chip and the
M) onto the preadsorbed
l
BSA surfaces. Table 1 summarizes the corresponding parameters
obtained from data analysis.
Stage (a) in Fig. 1 indicates that BSA adsorbs on the sensor chip
very fast, rising rapidly the surface mass and thickness within the
first few seconds after injecting BSA solution. Then the surface
mass and thickness reach plateaus in the stage (b) after rinsing
with phosphate buffer. The thickness of BSA layer drops from
26.36 to 2.15 nm and the mass from 6.42 to 1.50 ng. The decrease
in both surface mass and thickness of BSA layer indicates that the
binding of the BSA onto the surface might be loose due to its big
size and rigidity [24]. Meanwhile the average layer density and
RI augment and then level out in stage (b). We might speculate that
the bilayer of BSA forms during the adsorption, and the out layer is
looser than the inner layer. When rinsed the phosphate buffer, the
out layer of BSA is washed away with the buffer and the inner one
remains. So the density and RI increase by washing. Stage (c) rep-
resents the immobilization of EQ14-2-14. In the first few seconds
after injecting EQ14-2-14, a rapid rise in thickness and mass is
accompanied by a sharp decrease in average layer density. At this
time the majority of the EQ14-2-14 remains loosely coiled above
the BSA film, the hydrophobic tail chains of EQ14-2-14 inserts to
the underlying BSA surface and the head groups extends into the
aqueous phase. After the fast adsorption of EQ14-2-14, the rate
of deposition of EQ14-2-14 decreases, mass and thickness almost
unchanged while the film density rises rapidly with augmenting
the mount of EQ14-2-14. In this period more EQ14-2-14 molecules
become associated with complementary anion sites of BSA, the tail
of EQ14-2-14 slowly diffuses into the hydrophobic cavity of BSA
ML ¼ qLsL
ð1Þ
where ML is the mass (ng) and
s
L the layer thickness (nm) resolved
from the phase data. qL can be calculated from the following
equation:
qL
¼
qMðnL ꢁ nBÞ=ðnM ꢁ nBÞ
q
ð2Þ
where
L is the layer density (g/cm3),
q
M is the density of the depos-
ited material (g/cm3), nL is the RI of the layer resolved from the
phase data, nB is the RI of the bulk solution calculated from the cal-
ibration, and nM is the RI of the deposited material [22].
2.6. Molecular docking
The structure of EQ14-2-14 was drawn by ChemBioDraw 12
package. After being optimized using Gaussian03 programs, the
molecule was optimized in energy and geometry using minimiza-
tion protocol in Discovery Studio 2.5 (Accelrys, San Diego, CA). This
optimized structure was used to generate conformations and then
used for the docking studies.
Molecular docking was performed using Discovery Studio 2.5 to
obtain the protein-surfactant binding mode and binding energy.
The crystal structure of BSA (PDB ID: 4F5S) was based on the X-
ray structure from the Protein Data Bank (http://www.pdb.org/).
Docking study was conducted by CDOCKER module. The CDOCKER
is a molecular docking tool using a flexible receptor based on
CHARMm force field. It generates diverse ligand conformations
from different seeds with high temperature molecular dynamics.
Then, it orients the ligand conformations within the defined pro-
tein active site by translating the center of the ligand. Each orien-
tation is subjected to simulated annealing molecular dynamics and
sorted according to the interaction energy. Molecular docking of
EQ14-2-14 to BSA was performed using 2000 heating steps with
700 K target temperature and 5000 cooling steps with 300 K target
temperature. The top ten docking poses were saved and were cho-
sen for following analysis. The binding energies of the EQ14-2-14
toward BSA were calculated using Calculate Binding Energies pro-
tocol with default parameters, and that surfactant minimization
were performed by using the Generalized Born implicit solvent
model to consider the flexibility of residues within binding sites
of BSA [23].
Fig. 1. Adsorption of BSA (20 lM, pH 7.4) and EQ14-2-14(62.34 lM, pH 7.4) at the
2.7. Surface tension measurements
hydrophilic siliconoxynitride interface: (a) BSA associate onto the bare siliconoxy-
nitride surface, (b) plateaus of BSA after buffer washing, (c) immobilization of EQ14-
2-14 onto the preadsorbed BSA surface, and (d) plateaus of BSA/EQ14-2-14 after
buffer washing.
Surface tensions of the solutions with different concentration of
EQ14-2-14, BSA and EQ14-2-14/BSA mixture were measured by

50
G. Wu et al. / Journal of Molecular Structure 1045 (2013) 47–54
Table 1
force as the surfactant concentration is increased. The lager CMC
Key parameters obtained from DPI analysis.
value of the mixture in the presence of BSA indicates large extent
interaction between EQ14-2-14 and BSA and the formation of
BSA/EQ14-2-14 complex [28,29].
Layer
Thickness
(nm)
RI
Density (g/
cm3)
Mass
(ng)
BSA
Before
rinsing
After
rinsing
Before
rinsing
After
26.36
2.15
3.77
3.71
1.38 0.24
1.46 0.70
1.50 0.90
1.49 0.89
6.42
1.50
2.93
2.85
3.1.3. FTIR spectra
It is widely recognized that the frequencies of IR bands related
to the amide I, II and III vibrations, particularly amide I, are sensi-
tive to the change of secondary structure of proteins. The spectrum
of BSA in Fig. 3 is obtained by subtracting the absorption of D2O
from the spectrum of the BSA/D2O solution, and that of BSA/
EQ14-2-14 by subtracting the absorption of EQ14-2-14/D2O from
that of BSA/EQ14-2-14/D2O.
BSA/EQ14-
2-14
rinsing
It is reported that the amide I peak locates in the region of
1600–1700 cmꢁ1 (mainly C@O stretching) and amide II band at
1548 cmꢁ1 (CAN stretching coupled with NAH bending mode)
[30]. Fig. 3 manifests that the peak of amide I bands shifts from
1645.6 cmꢁ1 to 1649.8 cmꢁ1 and amide II from 1565.0 cmꢁ1 to
1567.3 cmꢁ1 in the presence of EQ14-2-14. The spectrum profile
of BSA deforms because the ratio of peak height (Ha/Hb) is 0.80
in BSA alone, but it is 2.18 in BSA and EQ14-2-14 mixture. The re-
sults indicate that the secondary structure of BSA is altered owing
to the interaction of BSA with EQ14-2-14 [31].
film by hydrophobic interaction, and the layer is compacted [25].
Finally, the layer diminishes very little (from 3.77 to 3.71 nm) by
buffer rinsing and so do the density and mass loss, indicating that
the binding between BSA and EQ14-2-14 is very strong [26].
Analyzing the data in Table 1 could conclude that the interac-
tion of BSA with EQ14-2-14 results in great increase in density,
mass and thickness of co-adsorption layer resistant to rinsing, indi-
cating a strong combination of BSA with EQ14-2-14.
A quantitative analysis of the protein secondary structure for
the free BSA and BSA/EQ14-2-14 has been also carried out by
means of self-deconvolution, the second derivative and Gaussian
curve-fitting. Before estimating the percentage content of each
secondary structure, the component bands of amide I (1600–
1700 cmꢁ1) are assigned according to literature [32]. The bands
1615–1637 cmꢁ1 and 1680–1692 cmꢁ1 are generally assigned to
b-sheet, 1638–1648 cmꢁ1 to random coil, 1649–1660 cmꢁ1 to
3.1.2. Surface tension measurement
It has been reported that the interactions between surfactants
and proteins in aqueous media give rise to the formation of associ-
ation structures, thereby modifying the solution and interfacial
properties [27]. Fig. 2 depicts the surface tension curves of EQ14-
2-14, BSA and BSA/EQ14-2-14 system. The plots specify that the
surface tension of the BSA, EQ14-2-14 and BSA/EQ14-2-14 solu-
tions decline with the concentrations, and the ccmc is 58.6 and
43.4 mN/m for the solutions of BSA and EQ14-2-14, respectively,
suggesting that the gemini is more capable than BSA in reducing
the surface tension of water. The critical micelle concentration
a
-helix, and 1660–1680 cmꢁ1 to b-turn structure. The fitting is
graphed in Fig. 4 and corresponding results are listed in Table 2.
Upon the interaction, a decrease of -helix is observed from
a
54.01% to 31.56% with an increase in b-sheet structure from
32.92% to 37.57%, b-turn structure from 5.20% to 9.11%, and in
(CMC) of EQ14-2-14 and BSA are found to be 79.39
lM and
random structure from 7.86% to 21.76%. The decreased
a-helix
2.09 M, indicating that BSA is more apt to form micelle in solution
l
structure and increased b-sheet and turn and random structure
suggest a partial unfolding of BSA structure. Due to the unfolding,
EQ14-2-14 might be tightly connected with BSA, resulting in the
augmented density, mass and thickness observed by DPI.
than EQ14-2-14. The mixture of BSA/EQ14-2-14 exhibits decreased
surface tension and presents two break points in the surface ten-
sion versus concentration profile, the first one is called the critical
aggregation concentration (CAC) and the second CMC, and they are
Fig. 5 specifies that the CH2 bands of free BSA located at 2962,
2925 and 2852 cmꢁ1 shift to 2958, 2921 and 2847 cmꢁ1 at the
molar ratio of BSA:EQ14-2-14 = 1:2, and the CH2 bands of free
BSA located at 2962, 2925, 2873, and 2852 cmꢁ1 shift to 2948,
2893, 2870 and 2826 cmꢁ1 at the molar ratio of BSA:EQ14-2-
14 = 1:4. The shifting asymmetric and symmetric CH2 stretching
vibrations of BSA indicate that the presence of hydrophobic inter-
actions between surfactant aliphatic tails and the hydrophobic re-
gion in BSA, the results are well agree with literatures [30,33–36],
which supports well the results by DPI.
9.45 lM and 115.30 lM, respectively. It is reported that at low sur-
factant concentration a first interaction starts due to electrostatic
force, and then protein interacts with surfactant via hydrophobic
3.2. Change of EQ14-2-14 in the surrounding of BSA
The interaction between molecules causes the electrons redis-
tribution essentially. Chemical shift in 1H NMR spectroscopy offers
information about the electron movement outside the hydrogen
nucleus and other vicinal hydrogen nuclei. NMR spin–lattice relax-
ation time describes the involvement of a particular part of ligand
in the binding with BSA. The changes in the chemical shifts can
suggest an unequal participation of particular elements of the li-
gand molecule in the formation of the complex. Lack of changes
points to an equal participation of the investigated functional
groups [37].
The 1H NMR spectra of EQ14-2-14 in the absence and presence
of BSA were recorded at higher concentration of EQ14-2-14
Fig. 2. Surface tension versus concentration plots for EQ14-2-14, BSA and BSA/
EQ14-2-14 system.

G. Wu et al. / Journal of Molecular Structure 1045 (2013) 47–54
51
Fig. 3. FTIR difference spectra of free BSA and BSA/EQ14-2-14 complex. (C(BSA) = 1.0 ꢂ 10ꢁ4 M, C(EQ14-2-14) = 4.0 ꢂ 10ꢁ4 M).
Fig. 4. Second derivative resolution enhancement and curve-fitted amide I region (1700–1600 cmꢁ1) of FTIR spectra for free BSA and with BSA with EQ14-2-14. Conditions:
C
BSA = 1.0 ꢂ 10ꢁ4 M; CEQ14-2-14 = 4.0 ꢂ 10ꢁ4 M; T = 298 K.
Fig. 5. The infrared spectra of protein CH2 symmetric and asymmetric stretching vibrations for BSA (0.10 mM) and the BSA/EQ14-2-14 complexes in the region 3000–
2800 cmꢁ1. The contribution of CH2 stretching vibrations of the surfactant was subtracted in the region.
Table 2
(10 mM) and lower concentration of BSA (0.5 mM) in order to get
better signal of EQ14-2-14. Fig. 6 shows the 1H NMR spectra of
Percentage of the secondary structure of BSA and BSA/EQ14-2-14.
EQ14-2-14/D2O and BSA/EQ14-2-14/D2O with a molar ratio of
[BSA]:[EQ14-2-14] = 1:150. The spectra for other molar ratios are
available in Figs. S7–S12.
Compound
a
-Helix (%)
b-Sheet (%)
b-Turn (%)
Random (%)
BSA
54.01
31.56
32.92
37.57
5.20
9.11
7.86
21.76
BSA/EQ14-2-14
The chemical shift (d) and relaxation time (T1) of the protons in
EQ14-2-14 change almost regularly with the mole ratio used. The
chemical shift is associated with the circulation electrons that sur-
round the nucleus, and the field commonly opposed to the external
applied field [38]. Fig. S13 and Table 3 specify that the chemical
shifts of all protons except for methyl move up-field with augment
of BSA, which indicates that the effective field of these protons in
EQ14-2-14 molecule is decreased by magnetic shielding effect

52
G. Wu et al. / Journal of Molecular Structure 1045 (2013) 47–54
Fig. 6. The 1H NMR spectrum of EQ14-2-14 in phosphate buffer-D2O (pH 7.4) and in the presence of BSA (molar ratio [BSA]:[EQ14-2-14] = 1:150).
Table 3
Chemical shift (d) of proton of EQ14-2-14 in the absence and presence of BSA.
Molar ratio (nBSA:nEQ14-2-14)
d (ppm)
H1
H2
H3
H4
H5
H6
H7
H8
H9
H10
–
9.3553
9.3489
9.3459
9.3430
9.3428
9.3337
9.3150
8.3444
8.3413
8.3412
8.3391
8.3376
8.3342
8.3160
9.1047
9.0983
9.0973
9.0947
9.0914
9.0900
9.0358
9.6154
9.6043
9.6019
9.5974
9.5950
9.5869
9.5500
4.5904
4.5787
4.5772
4.5770
4.5726
4.5695
4.5560
4.0119
4.0017
3.9991
3.9964
3.9923
3.9858
3.9585
4.8853
4.8748
4.8742
4.8689
4.8655
4.8571
–
2.0483
2.0361
2.0354
2.0297
2.0263
2.0146
1.9887
1.2584
1.2541
1.2525
1.2499
1.2421
1.2381
1.2142
0.6302
0.6353
0.6355
0.6349
0.6351
0.6405
0.6597
1:300
1:250
1:200
1:150
1:100
1:50
owing to BSA interfering, and moreover, it is also point to an in-
crease of electron density surrounding these protons [39]. p–p
molecule which is stabilized by the interaction with the protein
has a longer correlation time than the other parts of the molecule
[39].
stacking is responsible for the up-field shift proton signal and the
increased electron density between the electron-deficit pyridinium
rings of EQ14-2-14 and the electron-abundant pyrrole ring in Trp
residues of BSA, and the hydrophobic tail of EQ14-2-14 should be
in the hydrophobic cavities of BSA [40]. However, the down-field
shift signal for the protons of methyl could be explained by the
magnetic deshielding effects due to the decrease of electron den-
sity. According to this phenomenon we may guess that the methyl
group may be located in more polar microenvironment while the
pyridinium ring is close to Trp residue in subdomain II, which is
confirmed by molecular ducking in the following Section 3.3.
As shown in Table 4 and 1/T1 values of protons in EQ14-2-14
except protons of methyl group rise up with increasing BSA con-
centration, especially 1/T1of the hydrogen atoms attached with
pyridinium ring and in the spacer (H5 and H6), suggesting that
the distance between BSA and EQ14-2-14 diminishes and the
changed conformation in the complex due to the immobilization
of the functional group of EQ14-2-14. The decrease of correlation
times points to the limited motion of the methyl group in EQ14-
2-14 [31].
D
d value is governed by intermolecular proximity of protons to
3.3. Molecular modeling
the aromatic ring and can serve to detect the different environ-
ment of protons in the ligand [20] (Table S1). The data in Table 3
disclose that the chemical shifts of H3 and H4 move up-field by
0.0689 and 0.0654 ppm, respectively. Because H4 are located be-
tween the positive nitrogen and C@O in ester group, and H3 is next
to the C@O, the result indicates that N+ and the ester group are
quite close to Trp residues of BSA compared to other part of
The complementary applications of molecule modeling are em-
ployed to improve the understanding of the interaction of EQ14-2-
14 and BSA. The surfactant is constructed, geometrically optimized,
and docked to the structure of BSA with high resolution.
From the docking calculation, the conformer with minimum
binding energy is picked up from the 10 minimum energy con-
formers, and is shown in Fig. 7a. EQ14-2-14 binds deep within
the pocket at subdomain II A in the minimum energy conformer.
The pyridinium rings are next to Trp 213 and one of the end methyl
is approaching to the subdomain III. Fig. 7b specifies that the
surfactant molecule is surrounded by the hydrophobic and
negatively charge residues, namely, Lys-187, Arg-194, Leu-197,
Arg-198, Ser-201, Phe-205, Arg-208, Ala-209, Ala-212, Trp-213,
Arg-217, Glu-291, Lys-294, Val-342, Ser-343, Leu-346, Lys-350,
EQ14-2-14 molecule, which creates the chance for p–p stacking.
When EQ14-2-14 binds with BSA, spin–lattice interaction
should change the T1 value of hydrogen nuclei in EQ14-2-14 that
interact closely with BSA [41] (Table S2). The nuclear relaxation
rate, defined as 1/T1 (1/s), is proportional to the correlation time
sc if dipole–dipole interaction is the major process for relaxation.
Correlation time decreases with the increase of the distance be-
tween the ligand and the macromolecule. A part of the small

G. Wu et al. / Journal of Molecular Structure 1045 (2013) 47–54
53
Table 4
Spin–lattice relaxation rate (1/T1) of proton of EQ14-2-14 in the absence and presence of BSA.
Molar ratio (nBSA:nEQ14-2-14)
1/T1 (sꢁ1
)
H1
H2
H3
H4
H5
H6
H7
H8
H9
H10
–
0.4344
0.4160
0.4168
0.4346
0.4470
0.4541
–
0.3500
0.3625
0.3734
0.3776
0.4583
–
0.2345
0.2242
0.3043
0.3215
0.3289
0.3418
0.3442
0.4651
0.6234
0.6431
0.7027
0.7576
–
0.0803
0.0493
0.0516
0.0528
0.0531
0.0571
–
0.2378
0.1565
0.1621
0.1615
0.1714
0.1871
0.2571
0.1545
0.1436
0.1453
0.1488
0.1495
–
4.7149
4.2452
4.4842
4.5259
4.2934
4.5662
6.8547
2.3488
2.3576
2.5914
2.6314
2.6652
2.6900
4.2852
1.1542
1.2357
1.1784
1.1587
1.1537
0.8889
0.6557
1:300
1:250
1:200
1:150
1:100
1:50
–
–
–
Fig. 7. Molecular docking analysis results: (a) predicted orientation of the lowest docking energy conformation of EQ14-2-14 (stick rendered) in the subdomain IIA (binding
site I) of BSA, (b) hydrophobic amino acid residues surrounding bound EQ14-2-14, (c) one hydrogen-bonding interactions between Trp-213 and bound EQ14-2-14, and (d)
distance between bound EQ14-2-14 and Trp-134 and Trp-213 in minimum energy docking pose.
Asp-450, Ser-453, Glu-478, Ser-479, Leu-480, Val-481 and Asn-
482, inferring that the interaction between EQ14-2-14 and BSA is
mainly hydrophobic in nature. BSA has a net charge of ꢁ18 at pH
7.0 which is distributed as ꢁ10 in domain I, ꢁ8 in domain II and
0 in domain III of ellipsoidal BSA [42,43]. Therefore, electrostatic
interactions may occur between the cationic surfactant molecules
and the BSA. Also, hydrogen bonding could be taken into account
due to the presence of several the polar and negative charge resi-
dues in the proximity of the bound surfactant molecule. Within a
distance of 2.6–3.5 Å between donor and accepter atoms, one

54
G. Wu et al. / Journal of Molecular Structure 1045 (2013) 47–54
hydrogen bond is formed between the oxygen in a ester group of
EQ14-2-14 and the adjacent atom of Trp-213 (2.072 Å)
Appendix A. Supplementary material
N
(Fig. 7c). Hydrogen bonding suggests a decrease in hydrophilicity,
supporting well the hydrophobic interaction between BSA and
EQ14-2-14 from FTIR analysis. In addition, forming hydrogen
bonds moves H5 beneath the C@O and H6 vicinal Trp-213
(Fig. 7c), resulting an upper-field chemical shift of the hydrogen
atom, which is consistent very well with the results by 1H NMR.
For the methyl groups in EQ14-2-14, the molecular docking points
that one of methyl is near Lys-187 and the other vicinal Glu-478,
and the two amino residue with extra amino or carboxyl group
are responsible for down-field shift signal in 1H NMR. From the
docking simulation the calculated free energy change of binding
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.molstruc.2013.04.
003.
References
[1] A. Sidorowicz, C. Mora, S. Jabłonka, A. Poła, T. Modrzycka, D. Modrzyckaa, K.
Michalak, J. Mol. Struct. 744–747 (2005) 711.
[2] A.G. Ivšic, O. Jovic, R. Matešic-Puac, T. Hrenar, J. Mol. Struct. 929 (2009) 149.
[3] S. Deep, J.C. Ahluwalia, Phys. Chem. Chem. Phys. 3 (2001) 4583.
[4] A. Stenstam, D. Topgaard, H. Wennerstrom, J. Phys. Chem. B 107 (2003) 7987.
[5] C.T.L. Jr, K.A. Smith, T.A. Hatton, Biochemistry 44 (2005) 524.
[6] S.S. Madaeni, E. Rostami, Chem. Eng. Technol. 31 (2008) 1265.
[7] Z.J. Cheng, J. Lumin. 132 (2012) 2719.
[8] A. Zaragoza, J.A. Teruel, F.J. Aranda, A. Marqués, M.J. Espuny, A. Ortiz, Langmuir
28 (2012) 1381.
[9] M. Liu, D.Z. Sun, B. Li, L.Y. Zhu, J. Disper. Sci. Technol. 30 (2009) 459.
[10] A. Sułkowska, M. Maciazek, J. Rownicka, B. Bojko, D. Pentak, W.W. Sułkowski, J.
Mol. Struct. 834–836 (2007) 162.
[11] Z.Y. Yang, R.R. Tang, Z.F. Zhang, J. Mol. Struct. 1030 (2012) 19.
[12] J. Chen, X.Y. Jiang, X.Q. Chen, Y. Chen, J. Mol. Struct. 876 (2008) 121.
[13] L.J. Sheng, H. Ding, Z. Wang, G.W. Song, J. Mol. Struct. 979 (2010) 152.
[14] M.Y. Hu, X. Wang, H. Wang, Y. Chai, Y. He, G.W. Song, Luminescence 27 (2012)
204.
[15] X.J. Guo, X.W. Han, J. Tong, C. Guo, W.F. Yang, J.F. Zhu, B. Fu, J. Mol. Struct. 966
(2010) 129.
[16] Z.H. Hou, N.B. Zhou, B.H. He, Y. Yang, X.Y. Yu, J. Mol. Struct. 1004 (2011) 116.
[17] R. Sivakumar, S. Naveenraj, S. Anandan, J. Lumin. 131 (2011) 2195.
[18] D. Wu, G.Y. Xu, Y.J. Feng, Y.J. Wang, Y.Y. Zhu, Colloid Polym. Sci. 287 (2009)
225.
(D
G) for the binding of BSA and EQ14-2-14 is ꢁ116.338 kcal molꢁ1
,
indicating a spontaneous combination of BSA and EQ14-2-14. This
is also the reason for the increased thickness and density observed
by DPI test. The distances between tryptophan and the bound sur-
factant obtained from docking simulation were 4.619 Å and
27.107 Å from Trp-213 and Trp-134, respectively (Fig. 7d), mani-
festing EQ14-2-14 moving from the more polar hydrophilic region
to the much less polar hydrophobic region [44–46]. So the
p–p
stacking is possible between the electron-deficit pyridinium rings
of EQ14-2-14 and the electron-abundant pyrrole ring in Trp resi-
dues of BSA.
Generally, the results by the molecular docking are in good
agreement with those by DPI, FTIR and 1H NMR, and especially of-
fer reasonable explanation for the behavior of the surfactant tan-
gled by macromolecule, such as BSA.
[19] X.M. Zhou, Q. Yang, X.Y. Xie, Q. Hu, F.M. Qi, Z.U. Rahman, X.G. Chen, Dyes
Pigments 92 (2012) 1100.
[20] T.Q. Wu, Q. Wu, S.Y. Guan, H.X. Su, Z.J. Cai, Biomacromolecules 8 (2007) 1899.
[21] K. Karim, J.D. Taylor, D.C. Cullen, M.J. Swann, N.J. Freeman, Anal. Chem. 79
(2007) 3023.
4. Conclusions
[22] Q.Q. Wang, K.C. Tam, Langmuir 25 (2009) 9898.
[23] X.L. Han, F.F. Tian, Y.S. Ge, F.L. Jiang, L. Lai, D.W. Li, Q.L. Yu, J. Wang, C. Lin, Y.
Liu, J. Photochem. Photobiol. B. 109 (2012) 1.
New gemini surfactant is synthesized and its binding with BSA
has been evaluated by multi methods. Conclusions can be drawn
from above discussion.
[24] X.B. Zhao, F. Pan, P. Coffey, J.R. Lu, Langmuir 24 (2008) 13556.
[25] Y. Wang, J. Wang, F. Yang, X.R. Yang, Anal. Chem. 84 (2012) 924.
[26] T.J. Lane, W.R. Fletcher, M.V. Gormally, M.S. Johal, Langmuir 24 (2008) 10633.
[27] S.K. Mehta, Bhawna, K.K. Bhasin, AnilKumar, J. Colloid. Interf. Sci. 323 (2008)
426.
[28] D. Wu, G.Y. Xu, Y.H. Sun, H.X. Zhang, H.Z. Mao, Y.J. Feng, Biomacromolecules 8
(2007) 708.
[29] H.K. Mandal, A. Kundu, S. Balti, A. Mahapatra, J. Colloid. Interf. Sci. 378 (2012)
110;
(1) Analyzing by DPI and FTIR reveals that adding the gemini
surfactant causes changes in thickness, mass, density and
RI of BSA, induces
structure rising.
a-helix declining, b-anti, turn and random
X.N. Zhao, Y. Liu, L.Y. Niu, C.P. Zhao, Spectrochim. Acta A 94 (2012) 357.
[30] P. Bourassa, S. Dubeau, G.M. Maharvi, A.H. Fauq, T.J. Thomas, H.A. Tajmir-Riahi,
Eur. J. Med. Chem. 46 (2011) 4344.
[31] U. Katrahalli, V.K.A. Kalalbandi, S. Jaldappagari, J. Pharm. Biomed. Anal. 59
(2012) 102.
[32] H.C. Wang, X.H. Jiang, L.M. Zhou, Z.J. Cheng, W.M. Yin, M. Duan, P.L. Liu, X.M.
Jiang, J. Lumin. 134 (2013) 138.
[33] D.M. Charbonneau, H.A. Tajmir-Riahi, J. Phys. Chem. B 114 (2010) 1148.
[34] P. Bourassa, C.D. Kanakis, P. Tarantilis, M.G. Pollissiou, H.A. Tajmir-Riahi, J.
Phys. Chem. B 114 (2010) 3348.
[35] J.S. Mandeville, H.A. Tajmir-Riahi, Biomacromolecules 11 (2010) 465.
[36] S. Dubeau, P. Bourassa, T.J. Thomas, H.A. Tajmir-Riahi, Biomacromolecules 11
(2010) 1507.
[37] A. Sułkowska, M. Maciazek-Jurczyk, B. Bojko, J. Równicka, W.W. Sułkowski, J.
Mol. Struct. 891 (2008) 278.
[38] Z.S. Ji, H.Z. Yuan, M.L. Liu, J.M. Hu, J. Pharm. Biomed. Anal. 30 (2002) 151.
[39] M. Maciazek-Jurczyk, A. Sułkowska, J. Równicka-Zubik, B. Bojko, A. Szkudlarek-
Hasnik, M. Knopik, W.W. Sułkowski, J. Mol. Struct. 993 (2011) 302.
[40] Y.L. Jiang, Med. Chem. 16 (2008) 6406.
(2) In the presence of BSA, the resonance signals of protons
except the protons in methyl group in EQ14-2-14 shifts
up-field, indicating that the presence of
p–p stacking and
the hydrophobic tail of EQ14-2-14 should be in the hydro-
phobic cavities of BSA. The relaxation time decreases with
increasing BSA ratio in the solution, indicating that EQ14-
2-14 is getting closer with BSA.
(3) The results from molecular docking confirm the results
deduced by DPI, FTIR and 1H NMR. It specifies that the pos-
itively charged pyridinium ring of EQ14-2-14 embeds into
the subdomain II of BSA,
p–p stack between the electron-
deficient pyridinium ring and electron-rich Trp residue,
hydrophobic interaction and hydrogen bonding are crucial
factors in the combination.
[41] M. Maciazek-Jurczyk, A. Sułkowska, B. Bojko, J. Równicka-Zubik, W.W.
Sułkowski, Spectrochim. Acta A 82 (2011) 181.
[42] E. Ayranci, O. Duman, Protein Peptide Lett. 11 (2004) 331.
[43] E. Ayranci, O. Duman, Food. Chem. 84 (2004) 539.
[44] S.R. Feroz, S.B. Mohamad, N. Bujang, S.N.A. Malek, S. Tayyab, J. Agric. Food
Chem. 60 (2012) 5899.
[45] S. Ghosh, S. Jana, N. Guchhait, J. Phys. Chem. B 116 (2012) 1155.
[46] S. Jana, S. Dalapati, S. Ghosh, N. Guchhait, J. Photochem. Photobiol. A 231
(2012) 19.
Acknowledgements
The authors are very grateful for financial supports from the
Science and Technology Administration of Sichuan Province (Pro-
ject No. 2009JY0062) and from the National Natural Science Foun-
dation of China (21176201).