Accepted Manuscript Title: Binding of glutathione and melatonin to human serum albumin: A comparative study Author: Xiangrong Li Su Wang PII: DOI: Reference:

S0927-7765(14)00646-8 http://dx.doi.org/doi:10.1016/j.colsurfb.2014.11.023 COLSUB 6746

To appear in:

Colloids and Surfaces B: Biointerfaces

Received date: Revised date: Accepted date:

14-7-2014 13-10-2014 16-11-2014

Please cite this article as: X. Li, S. Wang, Binding of glutathione and melatonin to human serum albumin: A comparative study, Colloids and Surfaces B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014.11.023 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.

Binding of glutathione and melatonin to human serum albumin: A comparative study

a

ip t

Xiangrong Li a,*, Su Wang b Department of Chemistry, School of Basic Medicine, Xinxiang Medical University, Xinxiang,

General surgery, The Third Affiliated Hospital of Xinxiang Medical University, Xinxiang, Henan,

us

b

cr

Henan, 453003, PR China

453003, PR China

an

*Address correspondence to Xiangrong Li, Department of Chemistry, School of Basic Medicine, Xinxiang Medical University, Xinxiang, Henan, 453003, PR China

M

Postal address: Department of Chemistry, School of Basic Medicine, Xinxiang Medical University, 601 Jin-sui Road, Hong Qi District, Xinxiang, 453003, PR China

te

d

Tel: +86-373-3029128

Ac ce p

E-mail: [email protected]

1

Page 1 of 33

Abstract: Binding of glutathione and melatonin to human serum albumin (HSA) has been studied using isothermal titration calorimetry (ITC) in combination with UV-vis absorption spectroscopy,

ip t

Fourier transform infrared (FT-IR) spectroscopy, and circular dichroism (CD) spectroscopy. Thermodynamic investigations reveal that glutathione/melatonin binds to HSA is driven by favorable

cr

enthalpy and unfavorable entropy, and the major driving forces are hydrogen bond and van der Waals

us

force. For glutathione, the interaction is characterized by a high number of binding sites, which suggests that binding occurs by a surface adsorption mechanism that leads to coating of the protein

an

surface. For melatonin, one molecule of melatonin combines with one molecule of HSA and no more melatonin binding to HSA occurs at concentration ranges used in this study. The UV-vis absorption,

microenvironmental changes of HSA.

M

FT-IR, and CD spectroscopy suggest that glutathione and melatonin may induce conformational and

te

d

Keywords: Human serum albumin; Glutathione; Melatonin; Isothermal titration calorimetry; UV-vis

Ac ce p

absorption; Fourier transform infrared (FT-IR) spectroscopy; Circular dichroism (CD) spectroscopy

2

Page 2 of 33

1. Introduction Glutathione is one of the most prominent endogenous low-molecular-weight thiols found in

ip t

mammals and significantly in demand as a drug for therapeutic purpose because of multiple biological functions in various tissues and its involvement in many diseases and malnutrition [1]. It is a tripeptide

cr

composed of glutamic acid, cysteine and glycine, and has two characteristic structural features: a

us

glutamyl linkage and a sulfydryl group [2]. Glutathione is found in cells exists in two forms: one is a reduced form (GSH) and other is in oxidized form as glutathione disulfide (GSSG) [3]. In healthy

an

living cells, more than 90% of glutathione is found as GSH, which can be converted to the oxidized form (GSSG) during oxidative stress, and can be reverted to the reduced form by the action of the

M

enzyme glutathione reductase [4]. Thus, the important physiological function of GSH is an essential endogenous antioxidant that plays a central role in cellular defense against toxins and free radicals [5].

te

d

In such case, two GSH molecules form a molecule of oxidized glutathione (GSSG) via the formation of double-sulfur bond [1]. GSH is also an important cofactor in cell metabolism, differentiation,

Ac ce p

proliferation, and apoptosis [6]. A decreased GSH level is associated with aging and various diseases, including cardiovascular, inflammatory, immune, and neurodegenerative diseases [7]. Melatonin (N-acetyl-5-methoxytryptamine) is an endogenous neurohormone secreted primarily from the pineal gland in mammals [8]. The other organs and tissues including retina, the gastrointestinal tract,

lymphocytes, gut, ovary, testes, bone marrow and lens have been reported to produce it as well [9]. Melatonin can also be synthesized in non-mammalian vertebrates, invertebrates and in some organisms

including dinoflagellates, algae and bacteria and it even can be found in a variety of plants [10-12]. The presence of melatonin in such a variety of organisms suggests that this substance is phylogenetically highly conserved and plays an important role in the function and survival of organisms. Melatonin

3

Page 3 of 33

plays roles in numerous physiologic activities such as neurogenesis, immunomodulation, improving immune defense, regulating circadian rhythms and sleep, intervening in lipid metabolism, and

ip t

inhibiting cancer growth [9]. It has also been proposed as a natural antioxidant and potent free radical scavenger [13]. In contrast with usual antioxidants, melatonin is known as a suicidal antioxidant,

cr

because it does not contribute in a redox cycling. Once melatonin is oxidized, it cannot be reduced to

us

its initial state, because some irreversible end-products are formed through reaction of melatonin with free radicals [14].

an

Glutathione and melatonin represent the major antioxidants in plasma and act as a primary defense in the blood against free radical attack. However, to our knowledge, an accurate and full basic data for

M

clarifying the binding mechanisms of glutathione and melatonin to plasma proteins remain unclear. Serum albumin is the most abundant protein in blood plasma (~60%) and serves as a depot and

te

d

transport protein for numerous endogenous and exogenous compounds [15]. Knowledge of interaction mechanisms between these two antioxidants and serum albumin is very important for us to understand

Ac ce p

the pharmacokinetics and pharmacodynamics of them. First, the drug-serum albumin interaction plays a dominant role in the bioavailability of drugs because the bound fraction of drugs is a depot, whereas the free fraction of drugs shows pharmacological effects [16]. Second, drug distribution is mainly controlled by serum albumin, because most drugs circulate in plasma and reach the target tissues by binding to serum albumin [17,18]. If a drug is metabolized and excreted from the body too fast because of low protein binding, the drug won’t be able to provide its therapeutic effect. Alternatively, if a drug has high protein binding and is metabolized and excreted too slowly, it may increase the drug’s half-life in vivo and lead to undesired side effects [19]. Furthermore, very high affinity binding of a drug to serum albumin may prevent the drug from reaching the target at all, resulting in insufficient

4

Page 4 of 33

tissue distribution and efficacy. Third, the competition between two drugs for the binding sites on serum albumin may result in a decrease in binding and hence an increase of the concentration of the

ip t

free biologically active fraction of one or both of the drugs. Co-administration of two drugs increases the free concentration of the drug with the lower affinity to serum albumin [20]. In addition, these

cr

hydrophobic binding pockets enable serum albumin to increase the apparent solubility of the

us

hydrophobic drugs in plasma and modulate their delivery to the cells in vivo [21]. In a word, the absorption, distribution, metabolism, and excretion properties of a drug can be significantly affected as

an

a result of its binding to serum albumin.

Isothermal titration calorimetry (ITC), which measures directly the heat evolved during a reaction,

M

is the method of choice for obtaining thermodynamic information. This is because only ITC allows researchers to obtain directly the variations of enthalpy ΔH0 and of entropy ΔS0, as well as the

te

d

association constant K and the stoichiometry of binding n, for an association process [22]. Unlike other methods, ITC does not require chemical modification or immobilisation of reactants since heat of

Ac ce p

binding is a naturally occurring phenomenon [23,24]. This sets the technique apart from fluorescence methods that often require labeling or are specific to proteins that contain a fluorophore that is accessible to a quencher. ITC can also be applied to systems where the complex formed is insoluble. This is a distinct advantage over many solution based techniques, including capillary electrophoresis, where complex insolubility can be problematic [25]. The present study examines the thermodynamics of the binding of these two antioxidants to HSA using isothermal titration calorimetry (ITC), and the consequent conformational changes have been monitored using UV-vis absorption spectroscopy, FT-IR, and CD. The cooperativity displayed between glutathione and melatonin has long been thought to be key to the activity of both compounds in vivo.

5

Page 5 of 33

Glutathione is a hydrophilic antioxidant and melatonin is a hydrophobic antioxidant, thus, the binding mechanism of these two antioxidants interact with HSA may be different. In the study, the calorimetric

ip t

results are then coupled with spectroscopic observations to understand these mechanisms underlying these interactions.

cr

2. Materials and methods

us

2.1. Materials

HSA, glutathione and melatonin were purchased from Sigma-Aldrich Chemicals Company (USA).

an

Glutathione was directly dissolved in phosphate buffer solution of pH 7.40 (0.01 mol L-1 PBS), and melatonin was dissolved in 99.5% ethanol and then diluted with phosphate buffer solution of pH 7.40

M

(0.01 mol L-1PBS). The stock solutions of glutathione and melatonin were prepared and used immediately because of oxidation under light and air. Double distilled water was used to prepare

te

d

solutions. The HSA was dissolved in a phosphate buffer solution of pH 7.40 (0.01 mol L-1 PBS). The HSA stock solution was prepared by extensive overnight dialysis at 4°C against the buffer. The

Ac ce p

concentration of the HSA was determined on a TU-1810 spectrophotometer (Puxi Analytic Instrument Ltd., Beijing, China) using the extinction coefficient ε280=36600 mol-1 L cm-1 [26]. The pH was determined on a pHS-2C pH-meter (Shanghai DaPu Instruments Co., Ltd, Shanghai, China) at ambient temperature. Sample masses were accurately weighed on a microbalance (Sartorius, BP211D) with a resolution of 0.01mg. All other reagents were all of analytical reagent grade and were used as purchased without further purification. 2.2. Isothermal Titration Calorimetry Titration of HSA with glutathione/melatonin was performed using a Model Nano-ITC 2G biocalorimetry instrument (TA, USA) at 298 K. All these solutions were thoroughly degassed prior to

6

Page 6 of 33

the titrations to avoid the formation of bubbles in the calorimeter cell. The sample cell was loaded with the phosphate buffer (PBS, 0.01 mol L-1) or protein solution and the reference cell contained double

ip t

distilled water. In a typical experiment, buffered HSA solution was placed in the 950 μL sample cell of the calorimeter and glutathione/melatonin solution was loaded into the injection syringe. Injections

cr

were started after baseline stability had been achieved. Glutathione/melatonin was titrated into the

us

sample cell by means of syringes via 25 individual injections, the amount of each injection was 10 µL. The first injection of 10 µL was ignored in the final data analysis. The contents of the sample cell were

an

stirred throughout the experiment at 200 rpm to ensure thorough mixing. Raw data were obtained as a plot of heat (μJ) against injection number and featured a series of peaks for each injection. These raw

M

data peaks were transformed using the instrument’s software to obtain a plot of enthalpy change per mole of injectant (ΔH0, kJ mol-1) against molar ratio. Control experiments included the titration of

te

d

glutathione/melatonin solution into buffer, buffer into HSA, and buffer into buffer, controls were repeated for the same HSA concentration used. The last two controls resulted in small and equal

Ac ce p

enthalpy changes for each successive injection of buffer and, therefore, were not further considered in the data analysis [27]. Corrected data refer to experimental data after subtraction of the glutathione/melatonin into buffer control data. Estimated binding parameters were obtained from ITC data using NanoAnalyze software

provided by the manufacturer. Data fits were obtained using either the independent binding sites (single site) model or the multiple binding sites (two sites) model. For the independent binding sites model the analytical solution for the total heat measured (Q) is determined by the formula:

⎧⎪ 1 + [M ]nK − Q = VΔH 0 ⎨[L] + ⎪⎩

(1 + [M ]nK − [L]K )2 + 4 K [L] ⎫⎪ 2K

⎬ ⎪⎭

(1)

where V is the volume of the calorimeter cell, ΔH0 is enthalpy, [L] is ligand concentration, [M] is 7

Page 7 of 33

macromolecule concentration, n is the molar ratio of interacting species, and K is the equilibrium binding constant [28]. The analytical solution for Q in the multiple binding sites model is determined

⎧ n ΔH 0 K [L ] n ΔH 0 K [L ]⎫ Q = V [M ]⎨ 1 1 1 + 2 2 2 ⎬ 1 + K 2 [ L] ⎭ ⎩ 1 + K1[L]

ip t

by the formula

(2)

cr

where n1 and n2 are the molar ratios of interacting species, ΔH01 and ΔH02 are the enthalpies, and K1

us

and K2 are the equilibrium binding constants for each of the multiple binding sites [28]. The changes in free energy (ΔG0) and entropy (ΔS0) are calculated using the following equation [29]:

an

ΔG 0 = − RT ln K = ΔH 0 − TΔS 0

M

2.3. Absorbance measurements

(3)

UV–vis absorption spectra were recorded with a TU-1810 spectrophotometer (Puxi Analytic

d

Instrument Ltd., Beijing, China) equipped with 1.0 cm quartz cells at 298 K. Buffer (control) and

te

samples were placed in the reference and sample cuvettes, respectively.

Ac ce p

The absorption data were analyzed using the following equation [30-32] to estimate the binding constant Ka between HSA and glutathione/melatonin.

A0 εG εG 1 = + A − A0 ε H −G − ε G ε H −G − ε G K a [Q]

(4)

where A0 and A are the absorbance of HSA in the absence and presence of glutathione/melatonin, εG and εH−G are the absorption coefficients of HSA and its complex with glutathione/melatonin,

respectively. [Q] is the concentration of glutathione/melatonin, Ka is analogous to the binding constants at the corresponding temperature. Thus, the double reciprocal plot of A0/(A-A0) vs. 1/[Q] is linear and the binding constant (Ka) can be estimated from the ratio of the intercept to the slope. 2.4. FT-IR measurements

8

Page 8 of 33

FT-IR measurements were performed on an Avatar 360 E. S. P. FT-IR spectrometer (PerkinElmer). All spectra were taken using the ATR method with a resolution of 4 cm-1. The FT-IR spectra of HSA (2.0×10-4 mol L-1) in the absence and presence of glutathione/melatonin, were recorded in the range of

ip t

1700-1500 cm-1 at pH 7.40 phosphate buffer and 298 K. The molar ratio of glutathione/ melatonin to

cr

HSA was maintained at 1:1. The corresponding absorbance contributions of buffer and free

us

glutathione/melatonin solutions were recorded and digitally subtracted under the same condition. The secondary structure compositions of free HSA and its glutathione/melatonin complex were estimated

an

from the shape of the amide I band, located at 1650-1660 cm-1. Fourier self-deconvolution and second derivative resolution enhancement were applied to increase the spectral resolution in the region of

M

1700-1600 cm-1. The resolution enhancement resulting from self-deconvolution and the second derivative is such that the number and the position of the bands to be fitted are determined. In order to

te

d

quantify the area of the different components of the amide I contour, revealed by self-deconvolution and second derivative, a least-square iterative curve fitting was used to fit the Gaussian line shapes to

Ac ce p

the spectra between 1700 and 1600 cm-1. The curve-fitting analysis was performed using the Peak Fit software.

2.5. Circular dichroism (CD) measurements The CD measurements were carried out on a Jasco J-715 spectropolarimeter under constant

nitrogen flush. For measurements in the far-UV region (190-260 nm), a quartz cell with a path length of 0.2 cm was used. Three scans were accumulated with continuous scan mode and a scan speed of 200 nm min-1 with data being collected at 0.2 nm and response time of 2 s. The sample temperature

was maintained at 298 K. The protein concentration was fixed to 2.0×10−6 mol L-1 and the glutathione/melatonin concentration used was also 2.0×10−6 mol L-1. Corresponding blanks (without

9

Page 9 of 33

HSA) were recorded and subtracted from the sample spectra, and results were taken as CD ellipticity in mdeg.

α − helix(%) =

− MRE208 − 4000 × 100 33000 − 4000

(5)

ip t

The percentage of α-helix can be calculated using the following equation [33]:

cr

where MRE208 is the observed mean residue ellipticity (MRE) value at 208 nm, 4000 is the MRE of

us

the β-form and random coil conformation cross at 208 nm, and 33,000 is the MRE value of the pure α-helix at 208 nm.

Intensity of CD (mdeg) at 208nm 10C p nl

(6)

an

MRE208 =

for HSA) and l is the path length (0.2 cm).

te

3.1. ITC studies

d

3. Results and discussion

M

where Cp is the molar concentration of the protein (HSA), n the number of amino acid residues (585

Ac ce p

A representative calorimetric titration profile of 9×10-2 mol L-1 glutathione with 2×10-4 mol L-1 HSA at pH 7.40 and 298 K is shown in Figure1A1. The thermodynamic parameters for the interaction

of glutathione with HSA obtained from ITC are listed in Table 1. Each peak in the binding isotherm represents a single injection of the glutathione into the HSA solution. Figure 1A2 shows the plot of enthalpy change (ΔH0) against [glutathione]/[HSA] molar ratio. The data appear not to accurately fit

to the independent binding sites (single site) model, but to fit well to the more complex multiple binding sites (two sites) model. The solid smooth line in Figure 1A2 represents the best fit of the

experimental data using the multiple binding sites model. For data analyzed using the multiple binding sites model, the first site has the association constant K1 of 7.269×103 L mol-1, while the second site has the association constant K2 of 8.976×103 L mol-1. A cooperativity index, k, can be 10

Page 10 of 33

calculated which equals 4K1K2/(K1 + K2)2, where K1 and K2 are the two association constants [34]. The k value describing the binding of glutathione to HSA is 0.989. The cooperativity index k≈1

ip t

observed here reflecting the fact that under this condition the probability of occupying the second binding site is not diminished [34]. According to the multiple binding sites model, the stoichiometric

cr

binding number (n) and enthalpy change (ΔH0) of 57.3 and -23.38 kJ mol-1 for the first site and 3.61

us

and -330.1 kJ mol-1 for the second site, respectively. The free energy change (ΔG0) and entropy change (ΔS0) evaluated from Eq. 3 is -21.99 kJ mol-1 and -15.17 J mol-1 K-1 for the first site and -22.35

an

kJ mol-1 and -1.032×103 J mol-1 K-1 for the second site, respectively.

Melatonin (1×10-3 mol L-1) binding to HSA (1×10-4 mol L-1) was also studied and the

M

representative calorimetric titration profile is shown in Figure 1B1. The thermodynamic parameters for the interaction of melatonin with HSA obtained from ITC are also listed in Table 1. The independent

te

d

binding sites model is thought to be valid because the data exhibited a sigmoid curve shaped titration isotherm with an inflection point at a molar ratio of approximately 1:1 [27]. The solid smooth line in

Ac ce p

Figure 1B2 represents the best fit of the experimental data using the independent binding sites model with the stoichiometric binding number (n), association constant (K), and enthalpy change (ΔH0) of

1.07, 1.440×105 mol-1 L and -185.4 kJ mol-1, respectively. The free energy change (ΔG0) and the entropy change (ΔS0) evaluated from Eq. 3 is -29.16 kJ mol-1 and -523.6 J mol-1 K-1, respectively. The negative values of free energy (ΔG0) and enthalpy (ΔH0) support that the binding of

glutathione/melatonin to HSA is spontaneous and exothermic. Ross and Subramanian [35] have characterized the sign and magnitude of the thermodynamic parameter associated with various individual kinds of interaction which may take place in protein association process. The negative entropy (ΔS0) and negative enthalpy (ΔH0) values of glutathione-HSA system and melatonin-HSA

11

Page 11 of 33

system indicate that the major driving forces are hydrogen bond and van der Waals force. Negative enthalpy could be due to immobilization of ligand which is not sufficiently counteracted by liberation of bound water. By the Eq. 3, the change of Gibbs free energy (ΔG0) is the comprehensive embodiment

ip t

of the changes of enthalpy (ΔH0) and entropy (ΔS0). The binding of glutathione/melatonin to HSA is

cr

driven by favorable enthalpy and unfavorable entropy. The interactions between drugs and proteins

us

include mainly two kinds: specific and nonspecific. The “nonspecific binding” is usually used to represent drugs that bind to proteins based on physical adsorption rather than specific binding such as

an

“lock and key” interactions between drugs and proteins [36]. Furthermore, the “specific binding” is often used to mean strong binding in general, but the “nonspecific binding” is often but not always

M

weaker than the “specific binding” [37]. The specific or nonspecific binding is generally defined from the magnitude of the association constants [38]. The association constants between the glutathione and

te

d

HSA are lower compared to other strong protein-ligand complexes with binding constants ranging from 107-108 L mol-1 [39]. The relatively low binding affinities indicate that the binding of glutathione to

Ac ce p

HSA may be nonspecific binding. The significant difference between glutathione and melatonin is in the effective value of the stoichiometric binding number n. The interaction between glutathione and HSA is characterized by a high value of n, i.e., n1+n2 = 60.9. As seen from Figure 1A2, saturation of binding sites did not occur until a molar ratio of approximately 120:1 was reached. The values of n1 and n2 are too big to believe that these binding site numbers are physiologically relevant. Considering

results of some literatures suggesting that if the interaction is characterized by a high number of binding sites, which suggests that binding occurs by a surface adsorption mechanism that leads to coating of the protein surface [27,40]. It is feasible that this number (n1+n2=60.9) could be accommodated through a single layer of interaction/adsorption on the HSA surface. Thus, we come to

12

Page 12 of 33

the same conclusion that the binding of glutathione to HSA may be nonspecific binding. Whereas, for melatonin-HSA system, the value of the stoichiometric binding number n approximately equals to 1,

ip t

suggesting that one molecule of melatonin combines with one molecule of HSA and no more melatonin binding to HSA occurs at concentration ranges used in this study.

cr

3.2. Absorption spectroscopic studies

us

UV-vis absorption technique can be used to explore the structural changes of protein and to investigate protein-ligand complex formation [41]. The UV-vis absorption spectra of HSA in the and

presence

of

glutathione/melatonin

obtained

by

utilizing

the

mixture

of

an

absence

glutathione/melatonin and phosphate buffer at the same concentration as the reference solution are

M

shown in Figure 2. HSA has two absorption peaks, the strong absorption peak at about 213 nm reflects the framework conformation of the protein, the weak absorption peak at about 279 nm appears to be

te

d

due to the aromatic amino acids (Trp, Tyr, and Phe) [42]. With gradual addition of glutathione/melatonin to HSA solution, the intensity peak of HSA at 213 nm decreases with a red shift

Ac ce p

and the intensity of the peak at 279 nm has minimal changes. The results can be explained that the interaction between glutathione/ melatonin and HSA leads to the loosening and unfolding of the protein skeleton and decreases the hydrophobicity of the microenvironment of HSA [43]. The absorption data were analyzed to estimate the corresponding binding constants Ka using Eq. 4

(Figure 3A2 and B2). The estimated Ka value is (2.225 ± 0.474)×103 L mol-1 and (1.104 ± 0.387)×105 L

mol-1 for HSA-glutathione system and HSA-melatonin system, respectively. The binding constants Ka are of the same order of magnitude obtained by ITC methods. We have checked a lot of literature and found that the binding mechanism of melatonin to serum albumin was not reported in the literature and the report on glutathione to serum albumin interaction

13

Page 13 of 33

included only glutathione-capped quantum dots interaction with serum albumin. Q. Yang et al. have studied the binding of glutathione modified CdTe quantum dots (CdTe@glutathione) to HSA, the results obtained by fluorescence spectroscopy indicated that CdTe@glutathione QDs could quench the

ip t

intrinsic fluorescence of HSA and the binding constant was found to be 3.00×106 L mol-1 at 298 K [44].

cr

B. Yang et al. have studied the interaction of glutathione-capped CdTe quantum dots of different sizes

us

with HSA by spectroscopy and ITC. The binding constant decreases from 4.50×105 to 1.34×105 L mol-1 as the size decreases from 3.0 nm (QDs-572) to 2.8 nm (QDs-555). Meanwhile, the binding

an

stoichiometry for QDs-572 and QDs-555 at 298 K was determined to be 0.269 and 0.204, respectively [45]. L. Ding et al. have studied the interaction of L-glutathione capped ZnSe QDs with BSA. The

M

binding constant was found to be 1.83×105 L mol-1 [46]. The binding constants of glutathione-capped

te

ITC and UV-vis measurements.

d

quantum dots interaction with serum albumin are all higher than glutathione-HSA obtained by us from

3.3. FT-IR spectroscopic studies

Ac ce p

The FT-IR spectra can be used to directly analyze the effect of glutathione/melatonin on

secondary structures of HSA. FT-IR spectra of proteins exhibit a number of amide bands, which represent different vibrations of the peptide moiety. Among the amide bands of the protein, the amide I band (1700-1600 cm-1, mainly C=O stretch) and amide II band (1600-1500 cm-1, C-N stretch coupled

with N-H bending mode) both have a relationship with the secondary structure of protein [47].

However, the amide I band is more sensitive to the change of protein secondary structure than the amide II band [48]. The FT-IR spectra of free HSA and the difference spectra after binding with glutathione/melatonin in phosphate buffer solution were recorded (Figure 3A). Since there is no major spectral shifting for the protein amide I band at 1653 cm-1 (mainly C=O stretch) and amide II band at

14

Page 14 of 33

1546 cm-1 (C-N stretch coupled with N-H bending mode) upon interaction with the glutathione/melatonin,

their

intensities

remarkably

decreased

upon

interaction

with

ip t

glutathione/melatonin. It is important to note that the decrease in the intensity of the amide I band is due to the decrease of the proportion of protein α-helix structure, the result also suggests HSA changes

upon

the

glutathione/melatonin-HSA

interaction

[48-54].

cr

conformational

us

Glutathione/melatonin indeed exerts some influence on the polypeptide carbonyl hydrogen bonding network and finally the reduction of the protein α-helix structure, but the effect is rather weak. The

an

chemical interaction between glutathione/melatonin and HSA is not likely to be a covalent binding [49,55].

M

The infrared self-deconvolution with second derivative and curve-fitting procedures were used to determine HSA secondary structures in the absence and presence of glutathione and melatonin. The

te

d

component bands of the infrared amide I band are attributed as follows: 1615-1637 cm-1 to β-sheet, 1638-1648 cm-1 to random coil, 1649-1660 cm-1 to α-helix, 1660-1680 cm-1 to β-turn, and 1680-1692

Ac ce p

cm-1 to β-antiparallel structures, respectively [56]. A quantitative analysis of the protein secondary structure for the free HSA and its glutathione/melatonin complex was carried out, and the results are shown in Figure S1 and Figure 4B. Upon glutathione interaction, the α-helix decreases from 50.58%

(free HSA) to 48.59% (glutathione-HSA) and the β-antiparallel structure decreases from 3.51% (free

HSA) to 1.80% (glutathione-HSA). The reduction of α-helix and β-antiparallel is accompanied by an increase in β-sheet. The β-sheet increases from 25.14% (free HSA) to 30.30% (glutathione-HSA). However, a minor decrease of random coil structure is observed from 11.67% (free HSA) to 10.38% (melatonin-HSA) and the β-antiparallel structure decreases from 3.51% (free HSA) to 1.25% (melatonin-HSA), while an increase in β-turn structure from 9.10% (free HSA) to 12.63%

15

Page 15 of 33

(melatonin-HSA). The structural changes suggest important difference in the binding modes of glutathione compared to melatonin.

ip t

3.4. CD spectroscopic studies CD is a sensitive technique to monitor the conformational changes in the protein. The CD

cr

spectra of HSA in the absence and presence of glutathione/melatonin are shown in Figure 5. As

us

shown in Figure 5, CD spectra of HSA exhibited two negative bands at 208 and 222 nm, which is characteristic of the typical α-helix structure of protein [57]. The binding of glutathione/melatonin to

an

HSA causes only a decrease in negative ellipticity at all wavelengths of the far-UV CD without any significant shift of the peaks, which clearly indicates the changes in the protein secondary structure,

M

and a decrease of the α-helix content in protein [58]. The α-helix content of protein was calculated from Eqs.5 and 6. It can be calculated that the native HSA solution has 55.52 % of α-helix, while

te

d

α-helix content of HSA decreases to 49.52 % and 53.18 % with the addition of glutathione and melatonin in the mole concentration ratios of 1:1, respectively. The result suggests the occurrence of

Ac ce p

conformational change at the secondary structural level in the reaction between glutathione/melatonin and HSA [59].

4. Conclusions

In this study, we present detailed thermodynamic analysis for glutathione and melatonin binding to

HSA. They interact with HSA through two different binding mechanisms. Our work suggests that glutathione interacts with HSA through the multiple binding sites model, but melatonin interacts with HSA according to the independent binding sites model. Data from ITC experiments suggest that the binding of glutathione/melatonin to HSA is driven by favorable enthalpy and unfavorable entropy, and the major driving forces are hydrogen bond and van der Waals force. The obtained binding constants

16

Page 16 of 33

for glutathione with HSA are in the intermediate range, the relatively low binding affinities indicate that the binding of glutathione to HSA may be nonspecific binding. The value of the stoichiometric

ip t

binding number n determined by ITC suggests the binding of glutathione to HSA via multiple surface sites on the protein surface. For melatonin-HSA system, the stoichiometric binding number n

cr

approximately equals to 1, suggesting that one molecule of melatonin combines with one molecule of

us

HSA and no more melatonin binding to HSA occurs at concentration ranges used in this study. UV-vis absorption spectra find that the interaction between glutathione/melatonin and HSA decreases the

an

hydrophobicity of the microenvironment of HSA. The results of FT-IR and CD suggest that glutathione and melatonin indeed exerts some influence on the secondary structure of HSA.

M

Acknowledgements

This work was supported by the National Natural Science Foundation of China (21173071) and

References

te

d

the Research Fund for the Doctoral Program of Higher Education of China (20114104110002).

Ac ce p

[1] J. Chen, Z. He, H. Liu and C. Cha, Electrochemical determination of reduced glutathione (GSH) by applying the powder microelectrode technique, J. Electroanal. Chem. 588 (2006) 324–330. [2] Q. Wang, Y. Guan and M. Yang, Application of superparamagnetic microspheres for affinity adsorption and purification of glutathione, J. Magn. Magn. Mater. 324 (2012) 3300–3305. [3] H.B. Noh, P. Chandra, J.O. Moon and Y.B. Shim, In vivo detection of glutathione disulfide and oxidative stress monitoring using a biosensor, Biomaterials 33 (2012) 2600–2607. [4] A. Pastore, G. Federici, E. Bertini and F. Piemonte, Analysis of glutathione: implication in redox and detoxification, Clin. Chem. Acta. 333 (2003) 19–39. [5] L.Y. Niu, Y.S. Guan, Y.Z. Chen, L.Z. Wu, C.H. Tung and Q.Z. Yang, BODIPY-Based Ratiometric

17

Page 17 of 33

Fluorescent Sensor for Highly Selective Detection of Glutathione over Cysteine and Homocysteine, J. Am. Chem. Soc. 134 (2012) 18928−18931.

ip t

[6] N. Ballatori, S.M. Krance, R. Marchan and C.L. Hammond, Plasma membrane glutathione transporters and their roles in cell physiology and pathophysiology, Mol. Aspects Med. 30 (2009)

cr

13–28.

us

[7] S.H. Koo, J.S. Lee, G.H. Kim and H.G. Lee, Preparation, Characteristics, and Stability of Glutathione-Loaded Nanoparticles, J. Agric. Food Chem. 59 (2011) 11264–11269.

an

[8] N.P. Johns, J. Johns, S. Porasuphatana, P. Plaimee and M. Sae-Teaw, Dietary Intake of Melatonin

Agric. Food Chem. 61 (2013) 913−919.

M

from Tropical Fruit Altered Urinary Excretion of 6 Sulfatoxymelatonin in Healthy Volunteers, J.

[9] Y. Kaya, L. Sarıkcıoğlu, M. Aslan, C. Kencebay, N. Demir, N. Derin, D.N. Angelov and F.B.

te

d

Yıldırım, Comparison of the beneficial effect of melatonin on recovery after cut and crush sciatic nerve injury: a combined study using functional, electrophysiological, biochemical, and electron

Ac ce p

microscopic analyses, Childs Nerv Syst. 29 (2013) 389–401. [10] B. Vivien-Roels and P. Pevet, Melatonin: presence and formation in invertebrates, Experientia. 49 (1993) 642–647.

[11] B. Poeggeler, I. Balzer, R. Hardeland and A. Lerchl, Pineal hormone melatonin oscillates also in the dinoflagellate Gonyaulax polyedra. Naturwissenschaften 78 (1991) 268–269. [12] L.C. Manchester, B. Poeggeler, F.L. Alvares, G.B. Ogden and R.J. Reiter, Melatonin immunoreactivity in the photosynthetic prokaryote Rhodospirillum rubrum: Implications for an ancient antioxidant system, Cell. Mol. Biol. Res. 41(1995) 391–395. [13] A. Galano, On the direct scavenging activity of melatonin towards hydroxyl and a series of

18

Page 18 of 33

peroxyl radicals, Phys. Chem. Chem. Phys. 13 (2011) 7178–7188. [14] A. Esfandiar, O. Akhavan and A. Irajizad, Melatonin as a powerful bio-antioxidant for reduction

ip t

of graphene oxide, J. Mater. Chem. 21(2011) 10907–10914. [15] T. Xu, X. Guo, L. Zhang, F. Pan, J. Lv, Y. Zhang and H. Jin, Multiple spectroscopic studies on

cr

the interaction between olaquindox, a feed additive, and bovine serum albumin, Food Chem. Toxicol.

us

50 (2012) 2540–2546.

[16] Y. Zhang, S. Shi, X. Chen, W. Zhang, K. Huang and M. Peng, Investigation on the interaction

an

between ilaprazole and bovine serum albumin without or with different C-ring flavonoids from the viewpoint of food-drug interference, J. Agric. Food Chem. 59 (2011) 8499–8506.

M

[17] S. Tabassum, W.M. Al-Asbahy, M. Afzal, F. Arjmand and R.H. Khan, Interaction and photo-induced cleavage studies of a copper based chemotherapeutic drug with human serum albumin:

te

d

spectroscopic and molecular docking study, Mol. BioSyst. 8 (2012) 2424–2433. [18] F. Deng, C. Dong and Y. Liu, Characterization of the interaction between nitrofurazone and

Ac ce p

human serum albumin by spectroscopic and molecular modeling methods, Mol. BioSyst. 8 (2012) 1446–1451.

[19] M.L. Hall, W.L. Jorgensen and L. Whitehead, Automated ligand- and structurebased protocol for in silico prediction of human serum albumin binding, J. Chem. Inf. Model. 53 (2013) 907–922. [20] Z. Omidvar, A. Asoodeh and J. Chamani, Studies on the antagonistic behavior between cyclophosphamide hydrochloride and aspirin with human serum albumin: time-resolved fluorescence spectroscopy and isothermal titration calorimetry, J Solution Chem. 42 (2013) 1005–1017. [21] A.Y. Khan, M. Hossain and G.S. Kumar, Binding of plant alkaloids berberine and palmatine to serum albumins: a thermodynamic investigation, Mol. Biol. Rep. 40 (2013) 553–566.

19

Page 19 of 33

[22] D. Burnouf, E. Ennifar, S. Guedich, B. Puffer, G. Hoffmann, G. Bec, F. Disdier, M. Baltzinger and P. Dumas, kinITC: A new method for obtaining joint thermodynamic and kinetic data by isothermal

ip t

titration calorimetry, J. Am. Chem. Soc. 134 (2012) 559–565. [23] C. Poncet-Legrand, C. Gautier, V. Cheynier and A. Imberty, Interactions between Flavan-3-ols

cr

and Poly( L-proline) Studied by Isothermal Titration Calorimetry: Effect of the Tannin Structure, J.

us

Agric. Food Chem. 55 (2007) 9235–9240.

[24] Y. Liang, Applications of Isothermal Titration Calorimetry in Protein Folding and Molecular

an

Recognition, J. Iran. Chem. Soc. 3 (2006) 209–219.

Food Sci. Technol. 15 (2004) 186–190.

M

[25] A. Papadopoulou and R.A. Frazier, Characterization of protein–polyphenol interactions, Trends

[26] G.J. Zhang, B. Keita, C.T. Craescu, S. Miron, P.D. Oliveira and L. Nadjo, Polyoxometalate binding

11253–11259.

te

d

to human serum albumin: a thermodynamic and spectroscopic approach, J. Phys. Chem. B 111 (2007)

Ac ce p

[27] M.A. Dobreva, R.A. Frazier, I. Mueller-Harvey, L.A. Clifton, A. Gea and R.J. Green, Binding of

pentagalloyl glucose to two globular proteins occurs via multiple surface sites, Biomacromolecules 12 (2011) 710–715.

[28] E. Freire, O.L. Mayorga1 and M. Straume, Isothermal Titration Calorimetry, Anal. Chem. 62 (1990) 950A–958A.

[29] G. Wang, C. Yan and Y. Lu, Exploring DNA binding properties and biological activities of dihydropyrimidinones derivatives, Colloids Surf. B 106 (2013) 28–36. [30] J.J. Stephanos, Drug-protein interactions. Two-site binding of heterocyclic ligands to a monomeric hemoglobin. J. Inorg. Biochem. 62 (1996) 155–169.

20

Page 20 of 33

[31] W. Zhong, Y. Wang, J.S. Yu, Y. Liang, K. Ni and S. Tu, The interaction of human serum albumin with a novel antidiabetic agent-SU-118, J. Pharm. Sci. 93 (2004) 1039–1046.

ip t

[32] C.D. Kanakis, P.A. Tarantilis, M.G. Polissiou, S. Diamantoglou and H.A. Tajmir-Riahi, Antioxidant flavonoids bind human serum albumin, J. Mol. Struct. 798 (2006) 69–74.

cr

[33] A. Bhogale, N. Patel, J. Mariam, P. M. Dongre, A. Miotello and D. C. Kothari, Comprehensive

us

studies on the interaction of copper nanoparticles with bovine serum albumin using various spectroscopies, Colloids Surf. B 113 (2014) 276-284.

an

[34] G. Bains and E. Freire, Calorimetric determination of cooperative interactions in high affinity binding processes, Anal. Biochem. 192 (1991) 203–206.

M

[35] P.D. Ross and S. Subramanian, Thermodynamics of protein association reactions: forces contributing to stability, Biochemistry 20 (1981) 3096–3102.

te

d

[36] T. Tamura, T. Terada and A. Tanaka, A quantitative analysis and chemical approach for the reduction of nonspecific binding proteins on affinity resins, Bioconjugate Chem. 14 (2003)

Ac ce p

1222–1230.

[37] L.O. Andersson, A. Rehnstrom and D.L. Eaker, Studies on “nonspecific” binding, the nature of the binding of fluorescein to bovine serum albumin, Eur. J. Biochem. 20 (1971) 371–380. [38] A.D. Nielsen, K. Borch and P. Westh, Thermochemistry of the specific binding of C12 surfactants to bovine serum albumin, Biochim. Biophys. Acta. 1479 (2000) 321–331. [39] Z. Cheng, Studies on the interaction betweens copoletin and two serum albumins by spectroscopic methods, J. Lumin. 132 (2012) 2719–2729. [40] E.R. Deaville, R.J. Green, I. Mueller-Harvey, I. Willoughby and R.A. Frazier, Hydrolyzable tannin structures influence relative globular and random coil protein binding strengths, J. Agric. Food

21

Page 21 of 33

Chem. 55 (2007) 4554–4561. [41] Z.X. Chi and R.T. Liu, Phenotypic characterization of the binding of tetracycline to human serum

ip t

albumin, Biomacromolecules 12 (2011) 203–209. [42] X.C. Zhao, R.T. Liu, Z.X. Chi, Y. Teng and P.F. Qin, New insights into the behavior of bovine

cr

serum albumin adsorbed onto carbon nanotubes: comprehensive spectroscopic studies, J. Phys. Chem.

us

B 114 (2010) 5625–5631.

[43] X.R. Pan, P.F. Qin, R.T. Liu and J. Wang, Characterizing the interaction between tartrazine and

an

two serum albumins by a hybrid spectroscopic approach, J. Agric. Food Chem. 59 (2011) 6650–6656. [44] Q. Yang, X.M. Zhou, Y.S. Zhu and X.G. Chen, Study the interaction between CdTe@glutathione

M

and human serum albumin, J. Lumin. 135 (2013) 335–338.

[45] B. Yang, R. Liu, X. Hao, Y. Wu and J. Du, Effect of CdTe Quantum Dots Size on the

te

d

Conformational Changes of Human Serum Albumin: Results of Spectroscopy and Isothermal Titration Calorimetry, Biol. Trace Elem. Res. 155 (2013) 150–158.

Ac ce p

[46] L. Ding, P.J. Zhou, H.J. Zhan, C. Chen, W. Hu, T.F. Zhou and C.W. Lin, Microwave-assisted synthesis of L-glutathione capped ZnSe QDs and its interaction with BSA by spectroscopy, J. Lumin. 142 (2013) 167–172.

[47] J.N. Tian, J.Q. Liu, W.Y. He, Z.D. Hu, X.J. Yao and X.G. Chen, Probing the binding of scutellarin to human serum albumin by circular dichroism, fluorescence spectroscopy, FTIR, and molecular modeling method, Biomacromolecules 5 (2004) 1956–1961. [48] Y. Li, B. Jia, H. Wang, N. Li, G. Chen, Y. Lin and W. Gao, The interaction of 2-mercaptobenzimidazole with human serum albumin as determined by spectroscopy, atomic force microscopy and molecular modeling, Colloids Surf. B 104 (2013) 311–317.

22

Page 22 of 33

[49] Y.Q. Wang, X.Y. Wang, J. Wang, Y.M. Zhao, W.J. He and Z.J. Guo, Noncovalent interactions between a trinuclear monofunctional platinum complex and human serum albumin, Inorg. Chem. 50

ip t

(2011) 12661–12668. [50] D. Sarkar, Probing the interaction of a globular protein with a small fluorescent probe in the

cr

presence of silver nanoparticles: spectroscopic characterization of its domain specific association and

us

dissociation, RSC Adv. 3 (2013) 24389–24399.

[51] X. Guo, L. Zhang, X. Sun, X. Han, C. Guo and P. Kang, Spectroscopic studies on the interaction

an

between sodium ozagrel and bovine serum albumin, J. Mol. Struct. 928 (2009) 114–120. [52] Y. Yue, Y. Zhang, L. Zhou, J. Qin and X. Chen, In vitro study on the binding of herbicide

M

glyphosate to human serum albumin by optical spectroscopy and molecular modeling, J. Photochem. Photobiol. B 90 (2008) 26–32.

te

d

[53] P. Bourassa, I. Hasni and H.A. Tajmir-Riahi, Folic acid complexes with human and bovine serum albumins, Food Chem. 129 (2011) 1148–1155.

Ac ce p

[54] X.L. Guével, B. Hötzer, G. Jung, K. Hollemeyer, V. Trouillet and M. Schneider, Formation of Fluorescent Metal (Au, Ag) Nanoclusters Capped in Bovine Serum Albumin Followed by Fluorescence and Spectroscopy, J. Phys. Chem. C 115 (2011) 10955–10963. [55] Z.G. Peng, K. Hidajat and M.S. Uddin, Adsorption of bovine serum albumin on nanosized magnetic particles, J. Colloid Interface Sci. 271 (2004) 277–283. [56] G.W. Zhang, L. Wang and J.H. Pan, Probing the binding of the flavonoid diosmetin to human serum albumin by multispectroscopic techniques, J. Agric. Food Chem. 60 (2012) 2721–2729. [57] M. Bhattacharya, N. Jain and S. Mukhopadhyay, Insights into the mechanism of aggregation and fibril formation from bovine serum albumin, J. Phys. Chem. B 115 (2011) 4195-4205.

23

Page 23 of 33

[58] X. L. Han, F. F. Tian, Y. S. Ge, F. L. Jiang, L. Lai, D. W. Li, Q. L. Yu, J. Wang, C. Lin and Y. Liu, Spectroscopic, structural and thermodynamic properties of chlorpyrifos bound to serum albumin: A

ip t

comparative study between BSA and HSA, J. Photochem. Photobiol. B 109 (2012) 1-11. [59] S. Cao, B. Liu, Z. Li and B. Chong, A fluorescence spectroscopic study of the interaction between

Ac ce p

te

d

M

an

us

cr

Glipizide and bovine serum albumin and its analytical application, J. Lumin. 145 (2014) 94-99.

24

Page 24 of 33

Figure captions: Figure 1. The chemical structures of glutathione and melatonin. Figure 2. (A1) Raw data for the titration of 9×10-2 mol L-1 glutathione with 2x10-4 mol L-1 HSA at pH

ip t

7.40 and 298 K, showing the calorimetric response as successive injections of glutathione are added

cr

to the sample cell. (B1) Raw data for the titration of 1×10-3 mol L-1 melatonin with 1x10-4 mol L-1 HSA at pH 7.40 and 298 K, showing the calorimetric response as successive injections of melatonin are

us

added to the sample cell. (A2) Integrated heat profile of the calorimetric titration shown in panel A1.

an

The solid line represents the best nonlinear least-squares fit to the multiple binding sites model. (B2) Integrated heat profile of the calorimetric titration shown in panel B1. The solid line represents the

M

best nonlinear least-squares fit to the independent binding sites model.

Figure 3. UV-vis spectra of HSA in presence of different concentrations of glutathione (A1), and

d

melatonin (B1) at 298 K and pH 7.40. c(HSA)=2×10-6 mol L-1; c(glutathione)/(10-4 mol L-1) a-i: from

te

0.0 to 5.0. c(melatonin)/(10-5 mol L-1) a-g: from 0.0 to 2.6. The plots of A0/(A-A0) vs. [Q]-1 for

Ac ce p

glutathione-HSA system (A2), and melatonin-HSA system (B2) at 298 K and pH 7.40. Figure 4. (A) FT-IR spectra of free HSA and difference spectra [(glutathione-HSA)-glutathione], and [(melatonin-HSA)-melatonin] at pH 7.40; (B) Percentage of secondary structure motifs of the free HSA and its glutathione/melatonin complexes at pH 7.40. c(HSA) = 2.0×10-4 mol L−1; the molar ratio of glutathione/melatonin to HSA is 1:1. Figure 5. Circular dichroism spectra of HSA in the absence and presence of glutathione/melatonin. (a) 2.0×10-6 mol L−1 HSA, (b) 2.0×10-6 mol L−1 HSA in the presence of 2.0×10-6 mol L−1 melatonin and (c) 2.0×10-6 mol L−1 HSA in the presence of 2.0×10-6 mol L−1 glutathione.

25

Page 25 of 33

Table 1. Thermodynamic parameters for the interaction of glutathione and melatonin with HSA obtained from ITC. Melatonin-HSA

ΔH01/kJ mol-1

-23.38±4.95

-185.4±1.33

K1/L mol-1

(7.269±1.246)×103

(1.440±0.676)×105

n1

57.3±0.67

1.07±0.23

ΔG01/kJ mol-1

-21.99±0.43

ΔS01/J mol-1 K-1

-15.17±4.66

ΔH02/kJ mol-1

-330.1±4.89

/

K2/L mol-1

(8.976±3.488)×103

/

n2

3.61±0.79

/

-22.35±1.01

/

(-1.032±0.014)×103

/

cr

us -29.16±1.22 -523.6±0.30

an

M

d

te

ΔG02/kJ mol-1

Ac ce p

ΔS02/J mol-1 K-1

ip t

Glutathione-HSA

26

Page 26 of 33

ip t

Ac ce p

te

d

M

an

us

cr

Figure 1

27

Page 27 of 33

ip t cr us an M Ac ce p

te

d

Figure 2

28

Page 28 of 33

ip t cr us an M Ac ce p

te

d

Figure 3

29

Page 29 of 33

ip t cr us an M d te Ac ce p Figure 4

30

Page 30 of 33

ip t cr us an M Ac ce p

te

d

Figure 5

31

Page 31 of 33

us

cr

ip t

Graphical Abstracts

The study provides an accurate and full basic data for clarifying the different binding mechanisms of

Ac ce p

te

d

M

an

glutathione and melatonin to HSA.

32

Page 32 of 33

Research highlights

►Binding of glutathione to HSA occurs by a surface adsorption mechanism.

ip t

►The interaction between melatonin and HSA forms a 1:1 melatonin-HSA complex.

►Hydrogen bond and van der Waals force play a major role in the reaction.

Ac ce p

te

d

M

an

us

►Glutathione and melatonin may induce conformational changes of HSA.

cr

►Glutathione or melatonin binds to HSA is driven by favorable enthalpy and unfavorable entropy.

33

Page 33 of 33