Accepted Manuscript Binding interaction of sorafenib with bovine serum albumin: Spectroscopic methodologies and molecular docking Jie-Hua Shi, Jun Chen, Jing Wang, Ying-Yao Zhu PII: DOI: Reference:
S1386-1425(15)00506-5 http://dx.doi.org/10.1016/j.saa.2015.04.034 SAA 13585
To appear in:
Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy
Received Date: Revised Date: Accepted Date:
17 November 2014 12 March 2015 16 April 2015
Please cite this article as: J-H. Shi, J. Chen, J. Wang, Y-Y. Zhu, Binding interaction of sorafenib with bovine serum albumin: Spectroscopic methodologies and molecular docking, Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy (2015), doi: http://dx.doi.org/10.1016/j.saa.2015.04.034
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1
Binding interaction of sorafenib with bovine serum albumin: Spectroscopic
2
methodologies and molecular docking
3 4
Jie-Hua Shi a, b
*
Jun Chen a
Jing Wang a
Ying-Yao Zhu a
5
a
College of Pharmaceutical Science, Zhejiang University of Technology, Hangzhou 310032, China
6
b
State Key Laboratory Breeding Base of Green Chemistry Synthesis Technology, Zhejiang University of Technology,
7
Hangzhou 310032, China
8 9
Abstract:
The binding interaction of sorafenib with bovine serum albumin (BSA) was studied using fluorescence, circular
10
dichrosim (CD) and molecular docking methods. The results revealed that there was a static quenching of BSA induced by
11
sorafenib due to the formation of sorafenib–BSA complex. The binding constant and number of binding site of sorafenib with
12
BSA under simulated physiological condition (pH=7.4) were 6.8×104 M–1 and 1 at 310 K, respectively. Base on the sign and
13
magnitude of the enthalpy and entropy changes (ΔH0=-72.2 kJ mol-1 and ΔS0 =-140.4 J mol-1 K-1) and the results of molecular
14
docking, it could be suggested that the binding process of sorafenib and BSA was spontaneous and the main interaction
15
forces of sorafenib with BSA were van der Waals force and hydrogen bonding interaction. From the results of site marker
16
competitive experiments and molecular docking, it could be deduced that sorafenib was inserted into the subdomain IIA (site
17
I) of BSA and leads to a slight change of the conformation of BSA. And, the significant change of conformation of sorafenib
18
occurred in the binding process with BSA to increase the stability of the sorafenib-BSA system, implying that the flexibility
19
of sorafenib played an important role in the binding process.
20 21
Keywords: Sorafenib; Bovine serum albumin; Circular dichroism; Fluorescence spectroscopy; Molecular docking
22 23
*
Corresponding author at: College of Pharmaceutical Science, Zhejiang University of Technology, Hangzhou 310032, China. Tel./Fax: +86
571 8832 0064, E-mail address: [email protected] (Jie-hua Shi).
1
1
Introduction
2
Serum albumin is the most abundant carrier protein found in plasma with a high affinity for a wide range of
3
drugs and metabolites, which plays important physiological roles in transportation, distribution and metabolism
4
of many endogenous and exogenous ligands [1]. Generally, the weak binding of ligands with serum albumin
5
leads to a short lifetime or poor distribution of ligands, as the strong binding results in decreasing the
6
concentration of free ligands in plasma. Therefore, the binding interaction of ligands with serum albumin is
7
fundamental to the observed biological activities and the investigation of binding interaction between ligands
8
and serum albumin is the first step to clarify the detailed understanding of the pharmacology of drugs. However,
9
bovine serum albumin (BSA) and human serum albumin (HSA) are homologous proteins [2], which have been
10
extensively studied with different small molecules in the last years by using experimental methods and
11
theoretical calculation methods successfully [3-9]. Owing to the clear structure of BSA, low cost and highly
12
structural homology with HSA, BSA has been an important representative serum albumin in laboratory practice
13
for the interaction of drug and protein.
14
Sorafenib, which chemical name is 4-[4-[[4-chloro-3-(trifluoromethyl)phenyl]carbamoylamino]phenoxy]-
15
N-methyl-pyridine-2-carboxamide (Fig.1), is an oral multi-kinase inhibitor by inhibiting the activity of Raf
16
serine/threonine kinases and receptor tyrosine kinases involved in tumor growth and angiogenesis to suppress
17
tumor growth [10]. Until now, sorafenib is approved for the treatment of advanced renal cell carcinoma,
18
hepatocellular carcinoma [11], and radioactive iodine resistant advanced thyroid carcinoma [12]. Furthermore, it
19
has also been demonstrated that sorafenib has the preclinical and clinical activity against several other tumor
20
types such as prostate cancer, non small cell lung cancer and so on [13]. In the study of Chris H. Takimoto [14],
21
sorafenib showed a good safety profile and encouraging anti-tumor effect when coadministered with other agents
22
in patients with advanced solid tumors. Many studies have shown that sorafenib is highly bound in human 2
1
protein [15,16] (>99.7%), suggesting that sorafenib is not easy to displace from plasma proteins. Therefore, the
2
research on the binding interaction between sorafenib and serum albumin has significance to clarify the detailed
3
pharmacological mechanism of sorafenib.
4
The aim of this study was to elucidate the binding interaction of sorafenib with BSA and to obtain important
5
information about binding interaction mode, binding constant, the effect of binding interaction on conformations
6
of sorafenib and BSA. To study the binding interactions of sorafenib with BSA, fluorescence emission
7
spectroscopy, circular dichroism (CD), and molecular docking had been used in this work. This study is expected
8
to provide important insight into further elucidating the store and transport process of sorafenib in the body and
9
the mechanism of action and pharmacokinetics.
10
Materials and methods
11
Chemical and reagents
12
Bovine serum albumin (BSA) was purchased from Shanghai Bobo Biotechnology Co., Ltd., Sorafenib
13
Tosylate was provided from Nanjing Ange Pharmacetutical Co.,Ltd. Tris(hydroxymethyl) aminomethane (Tris)
14
(>99%) was obtained from Shanghai Bobo biotechnology Co., Ltd. Phenylbutazone (≥99.9%) was purchased
15
from Hubei Hengshuo Chemical Co., Ltd (Hubei, China). Ibuprofen (≥99.9%) was provided from Zhejiang
16
University of Technology. Other chemicals were of analytical reagent grade and were used without further
17
purification.
18
Tris-HCl buffer solution (pH=7.40) consisted of Tris (0.050 M) and was adjusted to pH=7.40 by 36% HCl
19
solution. The stock solution of sorafenib (3.0×10-5 M) was prepared in the mixture of ethanol (lower than 40%)
20
and water due to its low solubility in water. BSA was dissolved in Tris–HCl buffer solution containing 0.05 M
21
NaCl (pH=7.4). Both phenylbutazone and ibuprofen were dissolved in methanol with a concentration of 3.0×
22
10-3 M. Redistilled water was used throughout the work. All solutions were kept in the dark at 4℃. 3
1
In this work, the final concentration of ethanol in the testing solutions was controlled to be less than 2.4 %
2
to ensure ethanol won’t affect the properties of BSA.
3
UV spectra measurements
4
UV spectra of all BSA solutions in the absence and presence of sorafenib were recorded on a UV-1601
5
spectrophotometer (Shimadzu, Kyoto, Japan) equipped with 1 cm quartz cells from 200 to 350 nm at room
6
temperature. The corresponding solution of sorafenib was used as the reference solution.
7
Fluorescence measurements
8
All fluorescence spectra were performed on an F96S spectrofluorometer (Lengguang Technology Co., Ltd.,
9
Shanghai, China) with a 1cm quartz cell. The fluorescence emission spectra of solutions were recorded from 300
10
to 450 nm with the excitation wavelength at 289 nm with 10 nm slit widths for the excitation and emission at
11
three different temperatures (298, 303 and 310 K). The mixture solutions of sorafenib and BSA were prepared by
12
adding appropriate amounts of sorafenib and BSA solutions to 10 mL standard flask and diluting to 10 mL using
13
a Tris-HCl buffer solution (pH 7.4) to obtain a final sorafenib concentration from 0 to 1.8 μM with a constant
14
BSA content of 1.5 μM. Each solution was repeated three times and each spectrum was the average of three
15
scans.
16
In addition, the fluorescence intensity will be reduced if inner filter effect is exist, the presence of any
17
compound in fluorescence determination system which have ultraviolet absorption at excitation and emission
18
wavelength will be responsible for inner filter effect [17, 18]. In this work, all fluorescence intensities were
19
corrected for inner filter effect by using the following equation:
20
Fcor Fobs
A1 A2 10 2
(1)
21
where Fcor is the fluorescence intensity corrected, Fobs is the fluorescence intensity calculated in experiment, A1
22
and A2 are the sum of the absorbance of all components in fluorescence determination system at 289 nm and the 4
1
emission wavelength (λ nm), respectively.
2
Synchronous fluorescence measurement
3
The synchronous fluorescence spectra of the mixture solutions of sorafenib and BSA were recorded on
4
FluoroMax-4 (HORIBA Scientific, France) with different scanning intervals of Δλ at ambient temperature. The
5
Δλ (Δλ=λem–λex) values were set at 15 nm and 60 nm, respectively, which characterize the properities of tyrosine
6
and tryptophan, respectively.
7
Circular dichroism (CD) measurements
8
Circular dichroism spectra were recorded on JASCO J-815 Spectrophotometer with a 1.0 cm quartz cuvette
9
(Japan Spectroscopic Company, Tokyo, Japan) at ambient temperature, in which the scan range was from 200 to
10
350 nm with an interval of 1 nm at a scan rate of 100 nm min–1. In this work, the concentration of Tris-HCl was
11
diluted to 2.5 mM to avoid affecting of high concentration of chloride ion (Cl-1) on the sign of CD spectroscopy.
12
Each spectrum was the average of three scans and was corrected by corresponding buffer blanks.
13
Site marker competitive experiments
14
The site marker competitive experiments were carried out using the phenylbutazone (as site I marker) and
15
ibuprofen (as site II marker). And, the concentrations of BSA and site markers were set at 1.5 μM and 3.0 μM,
16
respectively, while the concentration of sorafenib gradually increased from 0 to 1.8 μM.
17
Molecular docking
18
The starting geometry of sorafenib (NCBI, CID 216239) was obtained from the PubChem Compound
19
Database (http://pubchem.ncbi.nlm.nih.gov/summary/summary.cgi?cid=216239&loc=ec_rcs). The structure of
20
sorafenib was first treated by semi-empirical theory at PM3 level and then was optimized by density functional
21
theory (DFT) at B3LYP/6-31+g(d,p) level using Gaussian 03 software until all eigenvalue of the Hessian matrix
22
were positive [19]. 5
1
The
crystal
structure
of
BSA
(P02769)
was
obtained
from
SWISS-MODEL
Repository
2
(http://swissmodel.expasy.org/repository/?pid=smr01&zid=async). The structure of BSA was optimized by
3
energy minimization in the water box (10 Å×10 Å×10 Å) with sodium ions to maintain charge neutrality using
4
the Charmm 27 force field and the conjugate gradients method (30000 cycles) implemented in NAMD 2.9
5
program from the Theoretical and Computational Biophysics Group (http://www.ks.uiuc.edu/Research/namd/).
6
The temperature, cutoff and time step were set at 310 K, 12.0 and 2.0 fs, respectively. The optimized structures
7
of sorafenib and BSA were used for the molecular docking calculations.
8
The docking software Autodock 4.2 along with the Autodock Tools [20] was used to simulate the binding
9
interaction between sorafenib and BSA. Firstly, the polar hydrogen atoms in BSA molecule were added [21].
10
Secondly, the partial atomic charges of BSA and sorafenib were calculated using Gasteiger-marsili [22] and
11
Kollman methods [23], respectively. Thirdly, the different conformers of sorafenib were generated by using
12
Lamarkian genetic algorithm (LGA) with the same parameters of each docking. For the recognition of the
13
binding sites on BSA, the grid centers were set at point (6.113, –11.010, 7.120), point (6.841, 2.436, –14.718)
14
and point (–1.216, –20.239, –6.738) in this work and the grid maps of dimensions 60 Å×60 Å×60 Å with a
15
grid-point spacing of 0.375 Å to ensure an appropriate docking size of sorafenib accessible. The number of
16
genetic algorithm run and the number of evaluations were set to 200 and 2.5 million, respectively. The other
17
docking parameters were default settings. Based on RMS cluster tolerance between structures, the complexes of
18
sorafenib with BSA were sorted into clusters. Finally, according to the Autodock scoring function, the
19
conformation of sorafenib–BSA complex with the lowest energy was selected as the most stable conformation of
20
sorafenib–BSA complex.
21
Results and discussion
22
UV absorption of BSA 6
1
UV absorption measurement is a very simple but effective method for exploring the structural change and
2
understanding the complex formation. The UV spectra of BSA in the presence of sorafenib were shown in Fig.1.
3
The results revealed that there were two absorption bands for the BSA solution in the presence of sorafenib. The
4
absorption band at near 210 nm reflected the framework comformation of BSA and weak absorption bands at
5
near 280 nm belonged to the π→π* transition of the aromatic amino acids such as Trp, Tyr and Phe. It was
6
obvious that the intensity of UV-absorption of BSA increased with the addition of sorafenib. However, red shift
7
of maximum peak of BSA at 210 nm and blue shift of maximum peak of BSA at 280 nm were also noticed
8
probably due to the formation of sorafenib–BSA complex.
9 10
Fig. 2
Fluorescence quenching of BSA
11
The fluorescence quenching measurement is a sensitive and effective method for investigating the binding
12
interaction of small molecules with proteins. BSA has two tryptophan residues [24] (Trp-237 and Trp-159) which
13
have strongest fluorescence intensity and very sensitive to the fluorophore microenvironment. Therefore, Trp
14
residue is widely used as an intrinsic probe to study the binding properties of ligand with BSA.
15
The fluorescence spectra of BSA upon the addition of varies concentration of sorafenib in physiological
16
condition (pH=7.4) were shown on Fig. 3. As shown in Fig. 3, the fluorescence intensity of BSA decreased
17
gradually with the increases of sorafenib, while the maximum emission wavelength and the shape of peak didn’t
18
change, indicating that there was a change in the surrounding environment of the fluorophores (Trp-159 and
19
Trp-237) due to interaction of BSA with sorafenib, resulting in the fluorescence quenching of BSA. Generally,
20
the fluorescence quenching mechanism of protein induced by ligand can be divided into the dynamic quenching
21
and static quenching. The dynamic quenching is caused by molecular collisions between ligand and fluorophore,
22
while the static quenching results from the formation of ground-state complex of fluorophores with quenchers 7
1
[25]. However, the dynamic quenching constant increases as the temperature increases because higher
2
temperature increases the value of diffusion coefficient and enhances the process of electron transfer, on the
3
contrary, the static quenching constant decreases with the increasing temperature due to decreasing the stability
4
of ground-state complex. In addition, the quenching mechanism can also be distinguished by the value of kq, the
5
maximum scattering collision quenching rate constant between quenchers and BSA is 2×1010 M-1 S-1 [26].
6
Therefore, the trend of change of quenching constant (Ksv) and quenching rate constant with temperature is
7
generally used to elucidate the fluorescence quenching mechanism of protein induced by quencher. The
8
fluorescence quenching constant (Ksv) and quenching rate constant (kq) can be calculated by the Stern–Volmer
9
equation [27]:
10
F0 1 K SV [Q] 1 k q 0 [Q] F
or
F0 F K SV [Q] kq 0[Q] F
(2)
11
where F0 and F are the fluorescence intensities in the absence and presence of quencher, respectively. Ksv is the
12
Stern–Volmer quenching constant and [Q] represents the concentration of quencher, kq is the quenching rate
13
constant of the biomolecule and τ0 is the average fluorescence lifetime of bimolecular without quencher, which is
14
about 6 ns for BSA.
15
Fig. 3.
16
The plots of (F0–F)/F against [Q]τ0 of sorafenib quencher with BSA at different temperatures were shown
17
in Fig. 4. The values of Ksv and kq at different temperature were calculated by slope and intercept of Fig. 4 and
18
were presented in Table 1. As shown in Table 1, the values of Ksv decreased with the increase of temperature and
19
the values of kq were greater than the maximum scattering collision quenching rate constant (2×1010 M-1 S-1) at
20
the studied temperature range, indicating that the fluorescence quenching mechanism of BSA induced by
21
sorafenib was the static quenching due to the formation of sorafenib-BSA complex.
22
Table 1
8
1 2
Fig. 4.
Binding constant and binding modes
3
When the quenching mechanism is confirmed as the static quenching, it can be assumed that there are
4
several independent binding sites (n) of ligand on BSA, the binding constant (Kb) and binding sites (n) can be
5
calculated by the following equation [28]:
6
log[( F0 - F )/F ] logKb n log[Q]
7
Where Kb is the binding constant and n is the number of binding sites. The values of Kb and n can be calculated
8
by the intercept and slope of the above equation log[(F0–F)/F] versus log[Q] as shown in Fig. 5 and the results
9
are presented in Table 2. From Table 2, it can be found that the Kb value is 6.8×104 M-1 at 310 K, suggesting that
10
the binding interaction of sorafenib with BSA is very strong and the concentration of free sorafenib in plasma is
11
lower. The values of n at the studied temperature range are near to 1, indicating that the existence of a single
12
binding site on BSA for sorafenib. In addition, the correlation coefficients (r2) at three different temperatures are
13
greater than 0.99, which indicated that the interaction of sorafenib with BSA almost exactly match the
14
site-binding model based on double-log equation.
(3)
15
Fig. 5.
16
Table 2.
17
Thermodynamic parameters and binding forces
18
In the binding process of protein with drugs, the binding forces mostly include hydrogen bonding
19
interaction, van der Waals forces, electrostatic interaction and hydrophobic interaction, which can be confirmed
20
according to the signs and magnitudes of the thermodynamic parameters such as Gibbs free energy change (ΔG0),
21
enthalpy change (ΔH0) and entropy change (ΔS0). However, the thermodynamic parameters in the binding
22
process can be calculated by the van’t Hoff equation:
9
1
ln Kb
H 0 S 0 RT R
(4)
2
G 0 RT ln K b
3
where R is the gas constant. Kb is the binding constant at corresponding temperatures. Based on the viewpoint of
4
Ross and Subramanian [29], both ΔH0 and ΔS0 are negative, suggesting that the main force is van der Waals
5
force and/or hydrogen bonding interaction. Both ΔH0 and ΔS0 are positive, the main interaction force is a
6
hydrophobic interaction. ΔH0 is almost zero and ΔS0 is positive, the main interaction force is an electrostatic
7
force. The van’t Hoff plot for the interaction of sorafenib with BSA was shown in Fig. 6. The thermodynamic
8
parameters were calculated and listed in Table 2.
9
(5)
Fig. 6.
10
From Table 2, the negative values of ΔG0 and ΔH0 revealed that the binding of sorafenib with BSA was
11
spontaneous and exothermic process. The values of ΔH0 and ΔS0 were -72.2 kJ mol-1 and -140.4 J mol-1 K-1
12
respectively, indicating that the main interaction forces of sorafenib with BSA were van der Waals forces and/or
13
hydrogen bonding interaction.
14
Conformational change of BSA induced by sorafenib
15
CD spectroscopy and synchronous fluorescence spectroscopy, two powerful methods for characterizing the
16
structure of protein [30, 31], were used in this work to investigate the secondary structure of BSA after binding
17
sorafenib to BSA.
18
The CD spectra of BSA in the absence and presence of sorafenib under physiological condition (pH= 7.4) at
19
room temperature were shown in Fig. 7. As shown in Fig. 7, the intensities of negative bands at near 208 and
20
222 nm, which contribute to the π→π* and n→π* transfer for the peptide bond of the α-helix, slightly increased
21
without the change of the position and shape of peak, suggesting that the BSA retains its secondary α-helical
22
structure after sorafenib binding to BSA. In addition to the visual observation from CD spectra, the α-helical 10
1
content of BSA is always used to analyze the change of secondary structure of BSA after ligand binding to BSA,
2
which can be calculated according to the following equation [32]:
3
MRE
4
α-helix (%)
5
where MRE is mean residue ellipticity, Cp is the molar concentration of BSA, n denotes the number of amino
6
acid residues (583 for BSA), l is the path length (1.0 cm), 4000 is the MRE of the β-form and random coil
7
conformation cross at 208 nm while 33,000 is the MRE value of a pure α-helix at 208 nm. It can be seen from the
8
Table 3 that the α-helical content of BSA slightly increases from 52.13% to 53.49%, with the concentration of
9
sorafenib increased from 0 to 1.2 µM, suggesting that the conformational change of BSA induced by sorafenib is
10
ObservedCD (m deg) 10C p nl
（6）
( MRE ) 208 4000 CD (m deg) 100 33000 4000
（7）
slight.
11
Fig. 7.
12
Table 3
13
However, the CD spectra on the near UV region from 240-350 nm can be representative for the tertiary
14
structure of protein. The character bands at near 290 and 305 nm belong to the fine bands of tryptophan (Trp)
15
residues, the character bands at near 275 and 282 nm contribute to the fine bands of tyrosine (Tyr) residues, and
16
the bands at near 255, 260 and 27 nm belong to the fine bands of phenylalanine (Phe) residues [33, 34]. From
17
Fig. 5, it can be found that the intensities of these character bands for Trp, Tyr and Phe residues increase with the
18
increase of the concentration of sorafenib, indicating that the tertiary structure of BSA has changed due to the
19
interaction of BSA with sorafenib.
20
Unlike the steady-state fluorescence spectroscopy, synchronous fluorescence spectroscopy can provide the
21
characteristic information about the microenvironment in a vicinity of disparate chromophores. When the
22
scanning intervals Δλ (Δλ=λem–λex) were set at 15 and 60 nm, respectively, the characteristic information for the 11
1
Tyr and Trp residues can be obtained. And, the shifts of the maximum emission wavelength represent the
2
alteration of the polarity of the microenvironment around Tyr or Trp residues [35]. The synchronous fluorescence
3
spectra of BSA upon the addition of sorafenib were recorded in Fig.8. From Fig.8, it can be found that the
4
fluorescence intensities of Tyr and Trp residues decrease with the increase of sorafenib concentration, which is
5
with the consistent of the steady-state fluorescence spectroscopy. And, the maximum emission wavelength
6
slightly shift ( There was slight change of the secondary structure of BSA due to binding sorafenib. > The flexibility of sorafenib plays an important role in increasing the sorafenib–BSA stability.
GRAPHIC ABSTRACT： It was confirmed that sorafenib binds to site I of BSA via van der Waals and hydrogen bonding interactions and forms 1:1 complex with it through spectroscopic methods and molecular doching.
O O 10
H
Sorafenib
-10
3 (Csorafenib=1.2 µM)
-20 -30 -40 -50 -60 -70 200
220
-54 -56 -58 1 -60 -62 3 -64 -66 -68 205 210 215 220 225 Wavelength / nm
240
260
0.5 0.0 1 -0.5 -1.0 3 -1.5 -2.0 -2.5 -3.0 240 260 280 300 320 340 Wavelength / nm
280
254.9 nm 262.1 nm 268.5 nm 273.1 nm 279.5 nm 290.7 nm 302.1 nm
H CD [medg]
H
1 (Csorafenib=0.0 µM)
0
CD [medg]
O
CD [medg]
CF3 Cl
300
320
340
Wavelength / nm
Fluorescence Intensity / a.u.
350
BSA
1 (Csorafenib=0.0 µM)
300 250
7 (Csorafenib=1.8 µM)
200 150 100 50
λex=289 nm
0 300 320 340 360 380 400 420 440 Wavelength / nm
Sorafenib - BSA complex
S1386-1425(15)00506-5 http://dx.doi.org/10.1016/j.saa.2015.04.034 SAA 13585
To appear in:
Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy
Received Date: Revised Date: Accepted Date:
17 November 2014 12 March 2015 16 April 2015
Please cite this article as: J-H. Shi, J. Chen, J. Wang, Y-Y. Zhu, Binding interaction of sorafenib with bovine serum albumin: Spectroscopic methodologies and molecular docking, Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy (2015), doi: http://dx.doi.org/10.1016/j.saa.2015.04.034
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1
Binding interaction of sorafenib with bovine serum albumin: Spectroscopic
2
methodologies and molecular docking
3 4
Jie-Hua Shi a, b
*
Jun Chen a
Jing Wang a
Ying-Yao Zhu a
5
a
College of Pharmaceutical Science, Zhejiang University of Technology, Hangzhou 310032, China
6
b
State Key Laboratory Breeding Base of Green Chemistry Synthesis Technology, Zhejiang University of Technology,
7
Hangzhou 310032, China
8 9
Abstract:
The binding interaction of sorafenib with bovine serum albumin (BSA) was studied using fluorescence, circular
10
dichrosim (CD) and molecular docking methods. The results revealed that there was a static quenching of BSA induced by
11
sorafenib due to the formation of sorafenib–BSA complex. The binding constant and number of binding site of sorafenib with
12
BSA under simulated physiological condition (pH=7.4) were 6.8×104 M–1 and 1 at 310 K, respectively. Base on the sign and
13
magnitude of the enthalpy and entropy changes (ΔH0=-72.2 kJ mol-1 and ΔS0 =-140.4 J mol-1 K-1) and the results of molecular
14
docking, it could be suggested that the binding process of sorafenib and BSA was spontaneous and the main interaction
15
forces of sorafenib with BSA were van der Waals force and hydrogen bonding interaction. From the results of site marker
16
competitive experiments and molecular docking, it could be deduced that sorafenib was inserted into the subdomain IIA (site
17
I) of BSA and leads to a slight change of the conformation of BSA. And, the significant change of conformation of sorafenib
18
occurred in the binding process with BSA to increase the stability of the sorafenib-BSA system, implying that the flexibility
19
of sorafenib played an important role in the binding process.
20 21
Keywords: Sorafenib; Bovine serum albumin; Circular dichroism; Fluorescence spectroscopy; Molecular docking
22 23
*
Corresponding author at: College of Pharmaceutical Science, Zhejiang University of Technology, Hangzhou 310032, China. Tel./Fax: +86
571 8832 0064, E-mail address: [email protected] (Jie-hua Shi).
1
1
Introduction
2
Serum albumin is the most abundant carrier protein found in plasma with a high affinity for a wide range of
3
drugs and metabolites, which plays important physiological roles in transportation, distribution and metabolism
4
of many endogenous and exogenous ligands [1]. Generally, the weak binding of ligands with serum albumin
5
leads to a short lifetime or poor distribution of ligands, as the strong binding results in decreasing the
6
concentration of free ligands in plasma. Therefore, the binding interaction of ligands with serum albumin is
7
fundamental to the observed biological activities and the investigation of binding interaction between ligands
8
and serum albumin is the first step to clarify the detailed understanding of the pharmacology of drugs. However,
9
bovine serum albumin (BSA) and human serum albumin (HSA) are homologous proteins [2], which have been
10
extensively studied with different small molecules in the last years by using experimental methods and
11
theoretical calculation methods successfully [3-9]. Owing to the clear structure of BSA, low cost and highly
12
structural homology with HSA, BSA has been an important representative serum albumin in laboratory practice
13
for the interaction of drug and protein.
14
Sorafenib, which chemical name is 4-[4-[[4-chloro-3-(trifluoromethyl)phenyl]carbamoylamino]phenoxy]-
15
N-methyl-pyridine-2-carboxamide (Fig.1), is an oral multi-kinase inhibitor by inhibiting the activity of Raf
16
serine/threonine kinases and receptor tyrosine kinases involved in tumor growth and angiogenesis to suppress
17
tumor growth [10]. Until now, sorafenib is approved for the treatment of advanced renal cell carcinoma,
18
hepatocellular carcinoma [11], and radioactive iodine resistant advanced thyroid carcinoma [12]. Furthermore, it
19
has also been demonstrated that sorafenib has the preclinical and clinical activity against several other tumor
20
types such as prostate cancer, non small cell lung cancer and so on [13]. In the study of Chris H. Takimoto [14],
21
sorafenib showed a good safety profile and encouraging anti-tumor effect when coadministered with other agents
22
in patients with advanced solid tumors. Many studies have shown that sorafenib is highly bound in human 2
1
protein [15,16] (>99.7%), suggesting that sorafenib is not easy to displace from plasma proteins. Therefore, the
2
research on the binding interaction between sorafenib and serum albumin has significance to clarify the detailed
3
pharmacological mechanism of sorafenib.
4
The aim of this study was to elucidate the binding interaction of sorafenib with BSA and to obtain important
5
information about binding interaction mode, binding constant, the effect of binding interaction on conformations
6
of sorafenib and BSA. To study the binding interactions of sorafenib with BSA, fluorescence emission
7
spectroscopy, circular dichroism (CD), and molecular docking had been used in this work. This study is expected
8
to provide important insight into further elucidating the store and transport process of sorafenib in the body and
9
the mechanism of action and pharmacokinetics.
10
Materials and methods
11
Chemical and reagents
12
Bovine serum albumin (BSA) was purchased from Shanghai Bobo Biotechnology Co., Ltd., Sorafenib
13
Tosylate was provided from Nanjing Ange Pharmacetutical Co.,Ltd. Tris(hydroxymethyl) aminomethane (Tris)
14
(>99%) was obtained from Shanghai Bobo biotechnology Co., Ltd. Phenylbutazone (≥99.9%) was purchased
15
from Hubei Hengshuo Chemical Co., Ltd (Hubei, China). Ibuprofen (≥99.9%) was provided from Zhejiang
16
University of Technology. Other chemicals were of analytical reagent grade and were used without further
17
purification.
18
Tris-HCl buffer solution (pH=7.40) consisted of Tris (0.050 M) and was adjusted to pH=7.40 by 36% HCl
19
solution. The stock solution of sorafenib (3.0×10-5 M) was prepared in the mixture of ethanol (lower than 40%)
20
and water due to its low solubility in water. BSA was dissolved in Tris–HCl buffer solution containing 0.05 M
21
NaCl (pH=7.4). Both phenylbutazone and ibuprofen were dissolved in methanol with a concentration of 3.0×
22
10-3 M. Redistilled water was used throughout the work. All solutions were kept in the dark at 4℃. 3
1
In this work, the final concentration of ethanol in the testing solutions was controlled to be less than 2.4 %
2
to ensure ethanol won’t affect the properties of BSA.
3
UV spectra measurements
4
UV spectra of all BSA solutions in the absence and presence of sorafenib were recorded on a UV-1601
5
spectrophotometer (Shimadzu, Kyoto, Japan) equipped with 1 cm quartz cells from 200 to 350 nm at room
6
temperature. The corresponding solution of sorafenib was used as the reference solution.
7
Fluorescence measurements
8
All fluorescence spectra were performed on an F96S spectrofluorometer (Lengguang Technology Co., Ltd.,
9
Shanghai, China) with a 1cm quartz cell. The fluorescence emission spectra of solutions were recorded from 300
10
to 450 nm with the excitation wavelength at 289 nm with 10 nm slit widths for the excitation and emission at
11
three different temperatures (298, 303 and 310 K). The mixture solutions of sorafenib and BSA were prepared by
12
adding appropriate amounts of sorafenib and BSA solutions to 10 mL standard flask and diluting to 10 mL using
13
a Tris-HCl buffer solution (pH 7.4) to obtain a final sorafenib concentration from 0 to 1.8 μM with a constant
14
BSA content of 1.5 μM. Each solution was repeated three times and each spectrum was the average of three
15
scans.
16
In addition, the fluorescence intensity will be reduced if inner filter effect is exist, the presence of any
17
compound in fluorescence determination system which have ultraviolet absorption at excitation and emission
18
wavelength will be responsible for inner filter effect [17, 18]. In this work, all fluorescence intensities were
19
corrected for inner filter effect by using the following equation:
20
Fcor Fobs
A1 A2 10 2
(1)
21
where Fcor is the fluorescence intensity corrected, Fobs is the fluorescence intensity calculated in experiment, A1
22
and A2 are the sum of the absorbance of all components in fluorescence determination system at 289 nm and the 4
1
emission wavelength (λ nm), respectively.
2
Synchronous fluorescence measurement
3
The synchronous fluorescence spectra of the mixture solutions of sorafenib and BSA were recorded on
4
FluoroMax-4 (HORIBA Scientific, France) with different scanning intervals of Δλ at ambient temperature. The
5
Δλ (Δλ=λem–λex) values were set at 15 nm and 60 nm, respectively, which characterize the properities of tyrosine
6
and tryptophan, respectively.
7
Circular dichroism (CD) measurements
8
Circular dichroism spectra were recorded on JASCO J-815 Spectrophotometer with a 1.0 cm quartz cuvette
9
(Japan Spectroscopic Company, Tokyo, Japan) at ambient temperature, in which the scan range was from 200 to
10
350 nm with an interval of 1 nm at a scan rate of 100 nm min–1. In this work, the concentration of Tris-HCl was
11
diluted to 2.5 mM to avoid affecting of high concentration of chloride ion (Cl-1) on the sign of CD spectroscopy.
12
Each spectrum was the average of three scans and was corrected by corresponding buffer blanks.
13
Site marker competitive experiments
14
The site marker competitive experiments were carried out using the phenylbutazone (as site I marker) and
15
ibuprofen (as site II marker). And, the concentrations of BSA and site markers were set at 1.5 μM and 3.0 μM,
16
respectively, while the concentration of sorafenib gradually increased from 0 to 1.8 μM.
17
Molecular docking
18
The starting geometry of sorafenib (NCBI, CID 216239) was obtained from the PubChem Compound
19
Database (http://pubchem.ncbi.nlm.nih.gov/summary/summary.cgi?cid=216239&loc=ec_rcs). The structure of
20
sorafenib was first treated by semi-empirical theory at PM3 level and then was optimized by density functional
21
theory (DFT) at B3LYP/6-31+g(d,p) level using Gaussian 03 software until all eigenvalue of the Hessian matrix
22
were positive [19]. 5
1
The
crystal
structure
of
BSA
(P02769)
was
obtained
from
SWISS-MODEL
Repository
2
(http://swissmodel.expasy.org/repository/?pid=smr01&zid=async). The structure of BSA was optimized by
3
energy minimization in the water box (10 Å×10 Å×10 Å) with sodium ions to maintain charge neutrality using
4
the Charmm 27 force field and the conjugate gradients method (30000 cycles) implemented in NAMD 2.9
5
program from the Theoretical and Computational Biophysics Group (http://www.ks.uiuc.edu/Research/namd/).
6
The temperature, cutoff and time step were set at 310 K, 12.0 and 2.0 fs, respectively. The optimized structures
7
of sorafenib and BSA were used for the molecular docking calculations.
8
The docking software Autodock 4.2 along with the Autodock Tools [20] was used to simulate the binding
9
interaction between sorafenib and BSA. Firstly, the polar hydrogen atoms in BSA molecule were added [21].
10
Secondly, the partial atomic charges of BSA and sorafenib were calculated using Gasteiger-marsili [22] and
11
Kollman methods [23], respectively. Thirdly, the different conformers of sorafenib were generated by using
12
Lamarkian genetic algorithm (LGA) with the same parameters of each docking. For the recognition of the
13
binding sites on BSA, the grid centers were set at point (6.113, –11.010, 7.120), point (6.841, 2.436, –14.718)
14
and point (–1.216, –20.239, –6.738) in this work and the grid maps of dimensions 60 Å×60 Å×60 Å with a
15
grid-point spacing of 0.375 Å to ensure an appropriate docking size of sorafenib accessible. The number of
16
genetic algorithm run and the number of evaluations were set to 200 and 2.5 million, respectively. The other
17
docking parameters were default settings. Based on RMS cluster tolerance between structures, the complexes of
18
sorafenib with BSA were sorted into clusters. Finally, according to the Autodock scoring function, the
19
conformation of sorafenib–BSA complex with the lowest energy was selected as the most stable conformation of
20
sorafenib–BSA complex.
21
Results and discussion
22
UV absorption of BSA 6
1
UV absorption measurement is a very simple but effective method for exploring the structural change and
2
understanding the complex formation. The UV spectra of BSA in the presence of sorafenib were shown in Fig.1.
3
The results revealed that there were two absorption bands for the BSA solution in the presence of sorafenib. The
4
absorption band at near 210 nm reflected the framework comformation of BSA and weak absorption bands at
5
near 280 nm belonged to the π→π* transition of the aromatic amino acids such as Trp, Tyr and Phe. It was
6
obvious that the intensity of UV-absorption of BSA increased with the addition of sorafenib. However, red shift
7
of maximum peak of BSA at 210 nm and blue shift of maximum peak of BSA at 280 nm were also noticed
8
probably due to the formation of sorafenib–BSA complex.
9 10
Fig. 2
Fluorescence quenching of BSA
11
The fluorescence quenching measurement is a sensitive and effective method for investigating the binding
12
interaction of small molecules with proteins. BSA has two tryptophan residues [24] (Trp-237 and Trp-159) which
13
have strongest fluorescence intensity and very sensitive to the fluorophore microenvironment. Therefore, Trp
14
residue is widely used as an intrinsic probe to study the binding properties of ligand with BSA.
15
The fluorescence spectra of BSA upon the addition of varies concentration of sorafenib in physiological
16
condition (pH=7.4) were shown on Fig. 3. As shown in Fig. 3, the fluorescence intensity of BSA decreased
17
gradually with the increases of sorafenib, while the maximum emission wavelength and the shape of peak didn’t
18
change, indicating that there was a change in the surrounding environment of the fluorophores (Trp-159 and
19
Trp-237) due to interaction of BSA with sorafenib, resulting in the fluorescence quenching of BSA. Generally,
20
the fluorescence quenching mechanism of protein induced by ligand can be divided into the dynamic quenching
21
and static quenching. The dynamic quenching is caused by molecular collisions between ligand and fluorophore,
22
while the static quenching results from the formation of ground-state complex of fluorophores with quenchers 7
1
[25]. However, the dynamic quenching constant increases as the temperature increases because higher
2
temperature increases the value of diffusion coefficient and enhances the process of electron transfer, on the
3
contrary, the static quenching constant decreases with the increasing temperature due to decreasing the stability
4
of ground-state complex. In addition, the quenching mechanism can also be distinguished by the value of kq, the
5
maximum scattering collision quenching rate constant between quenchers and BSA is 2×1010 M-1 S-1 [26].
6
Therefore, the trend of change of quenching constant (Ksv) and quenching rate constant with temperature is
7
generally used to elucidate the fluorescence quenching mechanism of protein induced by quencher. The
8
fluorescence quenching constant (Ksv) and quenching rate constant (kq) can be calculated by the Stern–Volmer
9
equation [27]:
10
F0 1 K SV [Q] 1 k q 0 [Q] F
or
F0 F K SV [Q] kq 0[Q] F
(2)
11
where F0 and F are the fluorescence intensities in the absence and presence of quencher, respectively. Ksv is the
12
Stern–Volmer quenching constant and [Q] represents the concentration of quencher, kq is the quenching rate
13
constant of the biomolecule and τ0 is the average fluorescence lifetime of bimolecular without quencher, which is
14
about 6 ns for BSA.
15
Fig. 3.
16
The plots of (F0–F)/F against [Q]τ0 of sorafenib quencher with BSA at different temperatures were shown
17
in Fig. 4. The values of Ksv and kq at different temperature were calculated by slope and intercept of Fig. 4 and
18
were presented in Table 1. As shown in Table 1, the values of Ksv decreased with the increase of temperature and
19
the values of kq were greater than the maximum scattering collision quenching rate constant (2×1010 M-1 S-1) at
20
the studied temperature range, indicating that the fluorescence quenching mechanism of BSA induced by
21
sorafenib was the static quenching due to the formation of sorafenib-BSA complex.
22
Table 1
8
1 2
Fig. 4.
Binding constant and binding modes
3
When the quenching mechanism is confirmed as the static quenching, it can be assumed that there are
4
several independent binding sites (n) of ligand on BSA, the binding constant (Kb) and binding sites (n) can be
5
calculated by the following equation [28]:
6
log[( F0 - F )/F ] logKb n log[Q]
7
Where Kb is the binding constant and n is the number of binding sites. The values of Kb and n can be calculated
8
by the intercept and slope of the above equation log[(F0–F)/F] versus log[Q] as shown in Fig. 5 and the results
9
are presented in Table 2. From Table 2, it can be found that the Kb value is 6.8×104 M-1 at 310 K, suggesting that
10
the binding interaction of sorafenib with BSA is very strong and the concentration of free sorafenib in plasma is
11
lower. The values of n at the studied temperature range are near to 1, indicating that the existence of a single
12
binding site on BSA for sorafenib. In addition, the correlation coefficients (r2) at three different temperatures are
13
greater than 0.99, which indicated that the interaction of sorafenib with BSA almost exactly match the
14
site-binding model based on double-log equation.
(3)
15
Fig. 5.
16
Table 2.
17
Thermodynamic parameters and binding forces
18
In the binding process of protein with drugs, the binding forces mostly include hydrogen bonding
19
interaction, van der Waals forces, electrostatic interaction and hydrophobic interaction, which can be confirmed
20
according to the signs and magnitudes of the thermodynamic parameters such as Gibbs free energy change (ΔG0),
21
enthalpy change (ΔH0) and entropy change (ΔS0). However, the thermodynamic parameters in the binding
22
process can be calculated by the van’t Hoff equation:
9
1
ln Kb
H 0 S 0 RT R
(4)
2
G 0 RT ln K b
3
where R is the gas constant. Kb is the binding constant at corresponding temperatures. Based on the viewpoint of
4
Ross and Subramanian [29], both ΔH0 and ΔS0 are negative, suggesting that the main force is van der Waals
5
force and/or hydrogen bonding interaction. Both ΔH0 and ΔS0 are positive, the main interaction force is a
6
hydrophobic interaction. ΔH0 is almost zero and ΔS0 is positive, the main interaction force is an electrostatic
7
force. The van’t Hoff plot for the interaction of sorafenib with BSA was shown in Fig. 6. The thermodynamic
8
parameters were calculated and listed in Table 2.
9
(5)
Fig. 6.
10
From Table 2, the negative values of ΔG0 and ΔH0 revealed that the binding of sorafenib with BSA was
11
spontaneous and exothermic process. The values of ΔH0 and ΔS0 were -72.2 kJ mol-1 and -140.4 J mol-1 K-1
12
respectively, indicating that the main interaction forces of sorafenib with BSA were van der Waals forces and/or
13
hydrogen bonding interaction.
14
Conformational change of BSA induced by sorafenib
15
CD spectroscopy and synchronous fluorescence spectroscopy, two powerful methods for characterizing the
16
structure of protein [30, 31], were used in this work to investigate the secondary structure of BSA after binding
17
sorafenib to BSA.
18
The CD spectra of BSA in the absence and presence of sorafenib under physiological condition (pH= 7.4) at
19
room temperature were shown in Fig. 7. As shown in Fig. 7, the intensities of negative bands at near 208 and
20
222 nm, which contribute to the π→π* and n→π* transfer for the peptide bond of the α-helix, slightly increased
21
without the change of the position and shape of peak, suggesting that the BSA retains its secondary α-helical
22
structure after sorafenib binding to BSA. In addition to the visual observation from CD spectra, the α-helical 10
1
content of BSA is always used to analyze the change of secondary structure of BSA after ligand binding to BSA,
2
which can be calculated according to the following equation [32]:
3
MRE
4
α-helix (%)
5
where MRE is mean residue ellipticity, Cp is the molar concentration of BSA, n denotes the number of amino
6
acid residues (583 for BSA), l is the path length (1.0 cm), 4000 is the MRE of the β-form and random coil
7
conformation cross at 208 nm while 33,000 is the MRE value of a pure α-helix at 208 nm. It can be seen from the
8
Table 3 that the α-helical content of BSA slightly increases from 52.13% to 53.49%, with the concentration of
9
sorafenib increased from 0 to 1.2 µM, suggesting that the conformational change of BSA induced by sorafenib is
10
ObservedCD (m deg) 10C p nl
（6）
( MRE ) 208 4000 CD (m deg) 100 33000 4000
（7）
slight.
11
Fig. 7.
12
Table 3
13
However, the CD spectra on the near UV region from 240-350 nm can be representative for the tertiary
14
structure of protein. The character bands at near 290 and 305 nm belong to the fine bands of tryptophan (Trp)
15
residues, the character bands at near 275 and 282 nm contribute to the fine bands of tyrosine (Tyr) residues, and
16
the bands at near 255, 260 and 27 nm belong to the fine bands of phenylalanine (Phe) residues [33, 34]. From
17
Fig. 5, it can be found that the intensities of these character bands for Trp, Tyr and Phe residues increase with the
18
increase of the concentration of sorafenib, indicating that the tertiary structure of BSA has changed due to the
19
interaction of BSA with sorafenib.
20
Unlike the steady-state fluorescence spectroscopy, synchronous fluorescence spectroscopy can provide the
21
characteristic information about the microenvironment in a vicinity of disparate chromophores. When the
22
scanning intervals Δλ (Δλ=λem–λex) were set at 15 and 60 nm, respectively, the characteristic information for the 11
1
Tyr and Trp residues can be obtained. And, the shifts of the maximum emission wavelength represent the
2
alteration of the polarity of the microenvironment around Tyr or Trp residues [35]. The synchronous fluorescence
3
spectra of BSA upon the addition of sorafenib were recorded in Fig.8. From Fig.8, it can be found that the
4
fluorescence intensities of Tyr and Trp residues decrease with the increase of sorafenib concentration, which is
5
with the consistent of the steady-state fluorescence spectroscopy. And, the maximum emission wavelength
6
slightly shift ( There was slight change of the secondary structure of BSA due to binding sorafenib. > The flexibility of sorafenib plays an important role in increasing the sorafenib–BSA stability.
GRAPHIC ABSTRACT： It was confirmed that sorafenib binds to site I of BSA via van der Waals and hydrogen bonding interactions and forms 1:1 complex with it through spectroscopic methods and molecular doching.
O O 10
H
Sorafenib
-10
3 (Csorafenib=1.2 µM)
-20 -30 -40 -50 -60 -70 200
220
-54 -56 -58 1 -60 -62 3 -64 -66 -68 205 210 215 220 225 Wavelength / nm
240
260
0.5 0.0 1 -0.5 -1.0 3 -1.5 -2.0 -2.5 -3.0 240 260 280 300 320 340 Wavelength / nm
280
254.9 nm 262.1 nm 268.5 nm 273.1 nm 279.5 nm 290.7 nm 302.1 nm
H CD [medg]
H
1 (Csorafenib=0.0 µM)
0
CD [medg]
O
CD [medg]
CF3 Cl
300
320
340
Wavelength / nm
Fluorescence Intensity / a.u.
350
BSA
1 (Csorafenib=0.0 µM)
300 250
7 (Csorafenib=1.8 µM)
200 150 100 50
λex=289 nm
0 300 320 340 360 380 400 420 440 Wavelength / nm
Sorafenib - BSA complex
