Binding interaction of sodium benzoate food additive with bovine serum albumin: multi-spectroscopy and molecular docking studies

Sodium benzoate (SB) is widely used as a preservative in food industry, and bovine serum albumin (BSA) is a major carrier protein similar to human serum albumin (HSA), the study of the binding between the two has great significance on human health. In this paper, we systematically investigated the binding of SB and BSA under the simulated physiological conditions combining with various common analytical methods, e.g., fluorescence, UV–vis absorption, synchronous fluorescence and circular dichroism (CD) spectra, as well as molecular docking method. The fluorescence quenching measurements were respectively carried out at 298 K, 303 K and 308 K using the Stern–Volmer method. The results reveal that ground state SB–BSA complex was formed within the binding constants from 2.02 × 104 to 7.9 × 103 M−1. Meanwhile, the negative values of ΔH0 (− 43.92 kJ mol−1) and ΔS0 (− 111.6 J mol−1 K−1) demonstrated that both the hydrogen binding interaction and van der Waals forces contributed to stabilizing the SB–BSA complex. The site marker competitive experiments show that the SB and BSA bound at site I. Furthermore, the experimental results of UV–vis absorption, synchronous fluorescence and CD spectra indicate that the binding of SB and BSA may change the conformation of BSA. In addition, the molecular docking experiment suggests that hydrogen bond was formed in the interaction between SB and BSA.


Introduction
As a recognized food-grade preservative, sodium benzoate (SB), whose structure is shown in Fig. 1, is widely used in the food, cosmetic, and pharmaceutical industries [1][2][3]. For example, in the food industry, SB is used in a variety of foods and beverages, such as salads, kimchi, carbonated beverages, jams, juices, and soy sauce, due to its effective inhibition of fungal and bacterial growth during storage [2,4]. In addition, SB is also applicable in clinically practice and can to treat various diseases, such as urea cycle disorders, liver disease, multiple sclerosis, and early Alzheimer's disease and Parkinson's disease [5,6]. Although SB as a preservative is generally recognized as safe (GRAS), its concentration is limited to 0.1% by the US Food and Drug Administration (FDA) [7]. In recent years, studies have shown that the organic form of SB is nontoxic, but its synthetic form is toxic to organisms at chronic doses [8]. Furthermore, it has been reported that SB may damage mitochondrial DNA [9]. Nevertheless, the results of these investigations remain controversial. The interaction between SB and biomacromolecules requires a more in-depth research.
The effect of SB on health can be explored by studying the interaction between SB and Serum albumin (SA). SA, the most abundant translocator protein in blood circulation [10,11], has several critical physiological functions, for example, the maintenance of the colloidal osmotic blood pressure, and the transportation of various Open Access BMC Chemistry *Correspondence: zhangxy26@mail.sysu.edu.cn; zhengy35@mail.sysu. edu.cn 1 State Key Laboratory of Optoelectronic Materials and Technologies, Sun Yat-sen University, Guangzhou 510275, China Full list of author information is available at the end of the article endogenous and exogenous compounds [12,13]. Bovine serum albumin (BSA) is not only widely used in biomedical and pharmaceutical applications, but also widely utilized as a ligand-biological model to study the interactions between small molecules and globular proteins, due to its high stability, low-cost, versatile ligand-binding properties, medical importance and high structural homology with human serum albumin (HSA) (approximately 76%) [14,15]. BSA is a single-chain globular protein consisting of 583 amino acid residues, forming 17 disulfide bonds [16]. In addition, BSA contains two major specific ligand-binding sites located in the hydrophobic cavities in sub-domains II-A and III-A, which are also known as Sudlow's site I and Sudlow's site II, respectively [17]. Meanwhile, previous works have demonstrated that the durability and toxicity of chemicals have huge influence on the structure of BSA because of interaction effect [18]. Moreover, several studies have shown that secondary structure of BSA changes upon binding with small molecules [19][20][21]. Therefore, investigating the interaction between BSA and chemicals, especially small molecules is of great significance.
In this work, the interaction between SB and BSA was studied by employing multi-spectroscopic methods and molecular docking (MD) modeling. Fluorescence spectroscopy was first used for understand the binding mechanism, and quenching mechanism, binding constant, mode and site were analyzed. Then, UV-vis absorption, synchronous fluorescence and CD spectra were employed to determine the conformation and structure of BSA when SB was binding. Furthermore, to further interpret the experimental findings, molecular docking modeling was used to explore the molecular graph of the binding interaction.

Chemicals and reagents
All chemicals used in this experiment, including BSA (lyophilized powder, Mw ≈ 66.2 kDa, Sigma) and sodium benzoate (Sinopharm), are of high purity. Thus, no further purification is required after purchase. PBS buffer solution (20 mM, pH = 7.40) was prepared with the ultrapure water (ρ = 18.2 MΩ cm). Then, BSA solution was prepared using PBS as a solvent for the spectral experiments. Additionally, all solutions were stored in the dark environment with a low temperature of 4 °C before using.

Measurements of the fluorescence spectra
Fluorescence experiments were taken out with the RF-5301 (Shimadzu, Japan) fluorescence spectrophotometer. Firstly, BSA solution (1 × 10 −6 M) was added to a quartz cell with size of 1.0 cm. An equal amount of BSA was added to the reference solution to eliminate the absorbance of BSA itself, and the absorbance of PBS buffer was subtracted through base line correction. Then, SB of different concentrations from 0 to 8 × 10 −6 M was gradually dropped into the BSA using a microsyringe. The transportation of various molecules and materials in the blood circulatory system are regulated by albumin. Some of the small molecules entered into the blood are reversibly bound to plasma proteins, forming binding molecules. While those do not bind are free molecules. When the concentration of free molecules decreases, some of the binding molecules dissociate into free molecules, thus, they are always in dynamic equilibrium [22]. In our experiment, SB and BSA were let stand for 5 min to reach dynamic equilibrium [23]. During the fluorescence measurement, the slit width was set at 10 nm/10 nm, the excitation wavelength was 285 nm, and the scanning range of the fluorescence emission spectrum was 300-450 nm.

Measurements of the synchronous fluorescence
The synchronous fluorescence of SB-BSA was measured using the same concentrations of the mixture solutions as in the fluorescence quenching measurements, but at room temperature. Spectra were recorded at Δλ = 15 nm and 60 nm, which showed the tyrosine residue and tryptophan residue characteristics of the BSA.

Measurements of UV-vis spectra
The UV-vis absorption spectra were obtained by a UV-3600 (Shimadzu, Japan) spectrophotometer. The concentration of BSA was kept at 1 × 10 −6 M, while the concentration of the added SB ranged from 0 to 8 × 10 −6 M, each time with an increase of 2 × 10 −6 M. The absorption spectra of the BSA between 230 and 330 nm were recorded in a 1 cm quartz absorption cell.

Measurements of CD spectra
CD spectra were recorded using the spectropolarimeter of JASCO J-810 and a 1.0 cm quartz absorption cell. Note that the spectrum of a cell only with PBS buffer solution was firstly measured as background signal to remove the influence of PBS buffer solution.

Molecular docking
Molecular docking simulation was employed to study the molecular interaction between BSA and SB using Auto Dock Vina, an open-source software with significantly fast dock running speed and high molecular docking accuracy [24]. To prepare the protein and ligand molecules for docking study, BSA crystal structure was first retrieved from the Protein Data Band (PDB ID: 3V03) (http://www.rcsb.org/struc ture/3V03), and then loaded on AutoDock Tools to remove additional molecules, e.g., all the water molecules. Next, Polar hydrogens and Gasteiger charges were added, respectively. The structure of SB was prepared by drawing the 2D chemical structure using ChemOffice and further optimized based on MM2 force field implemented using Chem3D. After that, the structurally optimized BSA and SB were employed to conduct molecular docking simulation. During the simulation, the size of the grid box along x-, y-and z-directions were all set at 18 Å and the grid spacing 1 Å. The grid box center was set at (88.537, 24.797, 13.111). The energy range and values were set at 4 kcal/mol and 100, respectively.

Results and discussion
Fluorescence spectral analysis of the interactions involving SB with BSA Fluorescence quenching of BSA Tryptophan (Trp), tyrosine (Tyr) and phenylalanine (Phe) residues are three main amino acids that can make protein generate endogenous fluorescence [25]. The major contribution to the changed fluorescence of BSA is from the environmentally-sensitive tryptophan (Trp) moiety [26]. The fluorescence emission spectra of BSA with various concentration of SB are shown in Fig. 2a. When excited at 285 nm, BSA had a characteristic band at around 344 nm. Furthermore, when the concentration of the added SB increased from 0 to 8 × 10 −6 M, the fluorescence intensity of BSA decreased significantly, indicating that the environment around the Trp residues of BSA varied with the addition of SB. Therefore, it can be inferred that there is a binding interaction between SB and BSA, and the binding site is located near the Trp residue [27].
The reaction temperatures for SB-BSA system were maintained at 298 K, 303 K, and 308 K, respectively. The fluorescence quenching data are analyzed by the Stern-Volmer equation [28]: where F and F 0 are the fluorescence intensity of BSA with and without the quencher, i.e., SB, respectively. K SV is the Stern-Volmer quenching constant with the unit being M −1 , and [Q] is the concentration of the quencher. k q is the quenching rate constant of BSA. τ 0 is the average fluorescence lifetime of BSA in the excited state without the quencher (the order of magnitude is 10 −8 ) [28]. K SV and K q value of BSA triggered by SB at different temperatures can be determined by calculating the slope of the curve, as shown in Fig. 2b. The values of the parameters K SV , k q and R at different temperatures are listed in Table 1. It can be seen from the results that K SV decreases from 1.87 × 10 4 to 1.34 × 10 4 M −1 as the temperature increases from 298 to 308 K. Moreover, the values of k q at various temperatures are all in the order of 10 12 M −1 s −1 , which , indicating that the SB-trigger BSA quenching process is static rather than dynamic.

Interaction parameters and binding model for SB-BSA complex
The binding constants (K b ) and binding site (n) of SB-BSA complex is calculated using the following formula [29]: According to the Eq. (2), K b and n can be calculated from the curve of log[(F 0 −F)/F] versus log[Q], as shown in Fig. 3a. The calculated results are summarized in Table 2. These results show that within the temperature range studied, the value n of SB-BSA complex is close to 1, indicating that BSA has a single high affinity binding site for SB. K b is calculated to be approximately 10 4 , indicating strong binding interactions between SB and BSA. It is also found that as temperature increases, K b value decreases, suggesting that the stability of SB-BSA complex decreases with the increasing of temperature.
To better understand the binding between BSA and SB, the van't Hoff Eq. (3) was used to calculate the thermodynamic enthalpy (H 0 ) and entropy (S 0 ) of BSA and SB complexation.
As shown in Fig. 3b, the curves of lnK b and 1/T were used to determine the thermodynamic parameters of SB-BSA complex at three different temperatures, i.e. 298 K, 303 K, and 308 K. Once the H 0 and S 0 values are determined, the variation in Gibbs free energy (G 0 ) can be calculated by the following standard Eq. (4).
Here, the binding constant values at three different temperatures, i.e. ΔG 0 , ΔH 0 and ΔS 0 , are listed in Table 2.   The negative value of ΔG 0 indicates that the interaction process between SB and BSA is spontaneous. And the positive H 0 and S 0 values indicate that hydrogen bonding and van der Waals interactions play a major role in the binding of the chemical to the protein [30,31].

Combination of fluorescent probes
To determine the displacement percentage of the fluorescent combination probe, according to the method introduced by Sudlow et al., the fluorescence markers for distinct binding sites were chosen: ketoprofen for site I and ibuprofen for site II [32]. And the following equation is also adopted.
where F 1 and F 2 represent the fluorescence intensity of the SB-BSA system in the absence and presence of the probe, respectively. As shown in Fig. 4. The increase in ketoprofen concentration results in significant decrease in the fluorescence intensity of the SB-BSA system. However, the increase in ibuprofen concentration has little impact on the fluorescence intensity of BSA. Therefore, SB and BSA are supposed to bind at site I [33].

Analysis of the UV-visible absorption spectra
UV absorption measurement, a very simple and effective method, is often used to observe the formation process and the change in the conformation of protein during the its binding interaction between with small molecule. The UV spectra of SB-BSA complex are shown in Fig. 5. All SB-BSA complex have absorption bands. The absorption band around 280 nm is the result of the π-π* transition of aromatic amino acids (Trp, Tyr, and Phe) [34]. Adding (5) Probe displacement (%) = F 2 /F 1 × 100 SB (0-8.0 × 10 −6 M) to the BSA solution enhances the absorption intensity (0.044-0.0556), and the maximum wavelength shows a slight blue shift at around 280 nm. The UV absorption intensity increases with the increase of SB concentration, indicating that complex is formed by SB and the amino acid residues of BSA.

Synchronous fluorescence spectroscopy
The synchronous fluorescence spectroscopy is a useful tool for obtaining information around chromophore microenvironment. Usually, the shift of the λem represents the alteration of the polarity of the amino acid environment (particularly Tyr or Trp residues) [35]. The synchronous fluorescence spectra were carried out for investigating structural change in BSA with SB addition (Fig. 6). The fluorescence spectra characteristic of tyrosine and tryptophan residues are shown in Fig. 6a (Δλ = 15 nm) and Fig. 6b (Δλ = 60 nm), respectively [36]. It is obvious that the fluorescence intensity of the tryptophan residues are much stronger than that of tyrosine residues in Fig. 6. And no significant change is found in the fluorescent emission peak position of both tyrosine and tryptophan residues. Thus, the addition of SB did not rearrange the microenvironment of the tyrosine and tryptophan residues in BSA.

Analysis of CD spectra
To further explore whether SB can induce conformational change in BSA, CD spectroscopy experiments were performed. The two negative bands at 208 nm (π-π*  transition) and 222 nm (n-π* transition) in CD spectrum of free BSA are characteristic of the protein α-helical structure [37], whose content can be estimated by [38]: where MRE 208 is the MRE value observed at 208 nm, 4000 is the MRE value of the β shape and random coil conformation at 208 nm, and 33000 is the MRE value of the pure α-helix at 208 nm. The MRE 208 value used to indicate the change in secondary structure of BSA determined by [38]: where n is the number of amino acid residues (583 for BSA). l is the cell path length, and C p is the molar concentration of BSA. When SB:BSA ratio increases from 0 to 2:1, the band intensities of the CD spectra have a slight decrease, whereas the peaks positions remain unchanged, as shown in Fig. 7. The results demonstrate that the secondary structure of BSA has a partial change from α-helical content. The α-helical content decreased from 62.7% of free BSA to 61.88% (SB:BSA = 1:1) and 60.24% (SB:BSA = 2:1). Based on these analyses, we conclude that the addition of SB alters the secondary structure of BSA, resulting in a decrease in α-helical content, which is nonetheless still dominant in the secondary structure.

Molecular docking
To further elucidate SB-BSA binding interaction, molecular docking was used to simulate the molecular interaction between BSA and SB. The simulated result of the predominate configuration of SB-BSA complex is plotted in Fig. 8, where the binding energy is the lowest. Moreover, as shown in Fig. 9, the binding results indicate that SB is very close to the amino acid residues Tyr149, Leu237, Arg256, Leu259, Ala260, Ile263, Ser286, Ile289, (6) a-helix (%) = [(− MRE 208 −4000) / (33000 −4000)] × 100 (7) MRE 208 = CD m deg 208 / 10 × n × l × C p and Ala290 at site I in subdomain IIA. The hydrogen bonding between SB and BSA (Tyr149, ARG256 with a bond length of 3.0 Å, 2.2 Å and 2.0 Å) is also responsible for maintaining the stability of the complex. Here, note that the calculated combined Gibbs free energy is − 5.7 kcal/mol (− 23.8 kJ/mol) and binding constant (K b ) is about 1.45 × 10 4 mol/L, which is quite different with our experimental result (ΔG = − 935 kJ/mol and K b = 2.2 × 10 4 mol/L). This may be caused by the difference between X-ray crystal structure of BSA and its solution state in the aqueous system.

Conclusion
In this study, we combined the multi-spectroscopic methods and molecular docking modeling to systematically investigate the interaction between SB and BSA. The multi-spectroscopic data show that the quenching of BSA binding with SB was caused by the formation of SB-BSA complex. The negative values of ΔH 0 and ΔS 0 demonstrated that hydrogen bonding and van der Waals forces contribute to make the SB-BSA complex stable. The negative values of ΔG indicate that the interaction process was spontaneous. Site marker competitive experiments show that SB and BSA bound at site I. In addition, the amino acid microenvironments and the secondary structure of BSA were altered by the addition of SB, as shown in UV-vis absorption, synchronous fluorescence spectroscopy and CD spectra data. Furthermore, molecular docking studies provided some valuable information on the interaction between SB and BSA and structural stability of their complex.
Abbreviations SB: sodium benzoate; BSA: bovine serum albumin; HSA: human serum albumin; CD: circular dichroism; GRAS: generally recognized as safe; FDA: Food and Drug Administration; SA: serum albumin; MD: molecular docking. Fig. 8 The docking conformation of SB-BSA complex and the lowest binding free energy Fig. 9 The interaction between SB and adjacent residues of BSA. The hydrogen bonds formed in the SB-BSA system are marked as yellow dotted lines