Interaction study of pioglitazone with albumin by fluorescence spectroscopy and molecular docking

Spectrochimica Acta Part A 78 (2011) 96–101

Contents lists available at ScienceDirect

Spectrochimica Acta Part A: Molecular and Biomolecular Spectroscopy journal homepage: www.elsevier.com/locate/saa

Interaction study of pioglitazone with albumin by fluorescence spectroscopy and molecular docking Farnoush Faridbod a , Mohammad Reza Ganjali b,∗ , Bagher Larijani b , Siavash Riahi c,a , Ali Akbar Saboury d , Morteza Hosseini e , Parviz Norouzi b , Christoph Pillip b a

Endocrinology and Metabolism Research Center, Tehran University of Medical Sciences, Tehran, Iran Center of Excellence in Electrochemistry, Faculty of Chemistry, University of Tehran, Tehran, Iran Institute of Petroleum Engineering, Faculty of Engineering, University of Tehran, P.O. Box 11365-456, Tehran, Iran d Institute of Biochemistry and Biophysics, University of Tehran, Tehran, Iran e Department of Chemistry, Islamic Azad University, Savadkooh Branch, Savadkooh, Iran b c

a r t i c l e

i n f o

Article history: Received 8 February 2010 Received in revised form 6 September 2010 Accepted 8 September 2010 Keywords: Fluorescence quenching Pioglitazone Human serum albumin Molecular docking study

a b s t r a c t Pioglitazone is a medicine of thiazolidinedione (TZD) class with hypoglycemic (antihyperglycemic, antidiabetic) action. Pioglitazone binding to human serum albumin (HSA) was investigated at different temperatures (290, 300 and 310 K) by fluorescence spectroscopic method. Molecular docking study was also carried out besides the experiments. Experimental results revealed that pioglitazone have an ability to quench the intrinsic fluorescence of HSA tryptophan through a static quenching procedure. The binding constant was determined using Stern–Volmer modified equation and energy transfer mechanisms of quenching were discussed. Thermodynamic parameters were also calculated according to enthalpy changes dependence on different temperatures. According to the theoretical and experimental results, hydrogen bonding was found to play a major role in the interaction of pioglitazone with HSA. © 2010 Elsevier B.V. All rights reserved.

1. Introduction Pioglitazone (Fig. 1) is a prescription drug of thiazolidinedione (TZD) class with hypoglycemic (antihyperglycemic, antidiabetic) action [1]. Pioglitazone reduces insulin resistance in the liver and peripheral tissues; increases the expense of insulin-dependent glucose; decreases withdrawal of glucose from the liver; reduces quantity of glucose, insulin and glycated haemoglobin in the bloodstream [2]. A drug’s efficiency may be affected by the degree to which it binds to the proteins within blood plasma [3]. Common blood proteins that drugs bind to are human serum albumin, lipoprotein, glycoprotein, ␣, ␤, and ␥ globulins [3]. Human serum albumin (HSA) is the most abundant blood plasma protein and is produced in the liver. HSA normally constitutes about 60% of human plasma protein. HSA concentrations in blood plasma range from 3.5 to 5.0 g/L [4]. Binding of a drug to albumin, results in an increased drug solubility in plasma, decreased toxicity, and protection against oxidation of the bound drug. Binding can also have a significant effect on the

∗ Corresponding author at: Center of Excellence in Electrochemistry, Faculty of Chemistry, University of Tehran, Tehran, Iran. Tel.: +98 21 61112788; fax: +98 21 66495291. E-mail address: [email protected] (M.R. Ganjali). 1386-1425/$ – see front matter © 2010 Elsevier B.V. All rights reserved. doi:10.1016/j.saa.2010.09.001

pharmacokinetics of drugs, e.g. prolonging in vivo half-life of the therapeutic agent [4]. However, too strong binding prevents drug release in tissues. That is why albumin binding information is one of the key characteristics of drug pharmacology. A drug in blood exists in two forms: bound and unbound. Depending on a specific drug’s affinity for plasma protein, a proportion of the drug may become bound to plasma proteins, with the remainder being unbound. If the protein binding is reversible, then a chemical equilibrium will exist between the bound and unbound states, such that: Protein + drug  Protein–drugcomplex

(1)

Notably, it is the unbound fraction of a drug which exhibits pharmacologic effects. It is also the fraction that may be metabolized and/or excreted. Since the unbound form is being metabolized and/or excreted from the body, the bound fraction will be released in order to maintain equilibrium. However, too strong binding prevents drug release in tissues. Many biological molecules, metal ions and a variety of drugs such as anticoagulants, tranquilizers, and general anesthetics (often more than 90% of the drug is bound) can bind to albumin in the blood and then be transported in the circulatory system [5]. HSA is a globular protein of 585 amino acids. HSA is composed of three structurally similar globular domains, each of which contains two subdomains, denominated subdomain IA, IB, IIA, IIB, IIIA, and IIIB. Each domain is suitable for binding a group of compounds. For

F. Faridbod et al. / Spectrochimica Acta Part A 78 (2011) 96–101

O

97

3. Results and discussion O S

3.1. Intrinsic HSA fluorescence

HN

N

O Fig. 1. Pioglitazone chemical structure.

example, two distinct metal-binding sites, one involving Cys-34 and the other the N terminus, make a total of six dominant areas of ligand association to albumin [4,6]. The effectiveness of drugs depends on their binding ability with albumins, so it is important to study the interactions between the drugs and this protein [7]. This study may provide thermodynamic information of pioglitazone binding to serum albumin, and become a useful research in pharmacokinetics, drug delivery, and clinical medicine. Fluorescence based protein assays offer many advantages over conventional techniques such as dialysis and ultra-filtration. Fluorescence spectroscopy offers a simple method without needing to separate the bound and unbound molecule. This reduces the time required for the experiment and eliminates the need for a size selective membrane. Also, dialysis and ultra-filtration require analysis of free and total drug concentration which can be resource and time consuming. Additionally, these methods cannot be used when the drugs bind extensively to the membrane; this is often a serious problem with highly hydrophobic drugs [8,9].

2. Experimental 2.1. Reagents Human serum albumin (HSA, 99%, fatty acid free, molecular weight of 66,478 Da) was purchased from Sigma Chemical Company (Sigma, St. Louis, USA) and Tris–HCl from Merck Co. Pioglitazone was purchased form Pharmacology Department of Tehran University of Medical Sciences. Triply distilled water was used during the experiments.

2.2. Apparatus The fluorescence studies were performed using a Perkin-Elmer LS50 spectrofluorimeter. The excitation wavelength was 295 nm for all cases with an excitation and emission band pass (slit) of 10 nm. Solutions were placed in a 1 cm path-length quartz cell for fluorescence measurements. All UV–Vis spectra were recorded in a UV–Vis Lambda 2 spectrophotometer (Perkin Elmer).

Fluorescence of the proteins is caused by three intrinsic fluorophores present in the protein structure; tryptophan, tyrosine and phenylalanine residues [11]. Because of very low quantum yield of phenylalanine and tyrosine, normally fluorescence of tryptophan residue is investigated in researches. Previous studies reveal that HSA has only one tryptophan residue, Trp-214 [11,12]. Generally, when the excitation wavelength of serum albumins was selected at 280 nm, the fluorescence was produced from tryptophan and tyrosine residues. At 295 nm, the emission is only due to the tryptophan residue [11,12]. According to the wavelength at which the maximum fluorescence intensity decreases we can find which residue is located in or near the binding position. In this study, tryptophan fluorescence emission spectra over the 300–400 nm wavelength range were recorded with the excitation set at 295 nm. Tryptophan is the dominant intrinsic fluorophore in proteins. In fact emission of HSA is dominated by tryptophan which absorbs at longer wavelength and displays the larger extinction coefficient [11]. 3.2. Fluorescence quenching of HSA Quenching measurements of albumin fluorescence is an important method to investigate the interactions of drugs and biological molecules with serum albumins [13–16]. A variety of molecular interactions can result in fluorescence quenching, including excited-state reactions, molecular rearrangements, energy transfer, ground-state complex formation, and collisional quenching [11]. The different mechanisms of fluorescence quenching are usually classified as either dynamic quenching or static quenching. Dynamic quenching and static quenching are caused by diffusion and ground-state complex formation, respectively. And they have different dependence on temperature [11]. Fig. 2 shows the fluorescence emission spectra (at ex = 295 nm, 300 K) of HSA during titration by pioglitazone. The maximum emission wavelength was near 331 nm. The same trends were observed at other temperatures. Fig. 2 also shows that the maximum HSA fluorescence intensity decreases near 331 nm. 3.3. Stern–Volmer quenching constant The quenching can be mathematically expressed by the Stern–Volmer equation (2), which allows for calculating of quenching constants. F0 = 1 + kq 0 [Q ] = 1 + KSV [Q ] F

where F0 and F are the fluorescence intensities in the absence and presence of quencher, kq is the bimolecular quenching constant,  0 is the lifetime of the fluorescence in the absence of quencher, [Q] is the concentration of the quencher, and KSV is the Stern–Volmer quenching constant. Obviously, KSV = kq 0

2.3. Sample treatment 2 × 10−5 mol L−1 HSA solutions were prepared in a Tris–HCl buffer [0.05 mol L−1 Tris base (2-amino-2-(hydroxymethyl)-1,3propanediol), 0.10 mol L−1 NaCl with a pH of 7.4 and stored at 0–4 ◦ C as stock solution. However, HSA can be safely used in the absence of buffers, due to the fact that albumin itself is a good buffer at neutral pH values [10]. The stock solution of pioglitazone (1.0 × 10−3 mol L−1 ) was also prepared in distilled water.

(2)

Because the fluorescence lifetime of the HSA is l0−8 s−1

(3)

[13], KSV can be determined by linear regression of a plot of F0 /F vs. [Q]. The Stern–Volmer plot of the titrations of HSA by pioglitazone is shown in Fig. 3. These plots show a small downward concave toward the x-axis at pioglitazone concentrations higher than 1.0 × 10−6 mol L−1 . At different temperatures Stern–Volmer plots have the same trends. At concentrations of pioglitazone lower than 1.0 × 10−6 mol L−1 , linear plots were produced with correlation coefficients, R2 = 0.995 for 290 and 300 K, and 0.997 for 310 K.

98

F. Faridbod et al. / Spectrochimica Acta Part A 78 (2011) 96–101 700.00

0.018 0.016

0.012

(F0-F)-1

Fluorescence Intensity (a.u.)

0.014 500.00

400.00

310 K

y = 0.0413x + 0.0032 R2 = 0.9985

600.00

y = 0.0213x + 0.0031 R2 = 0.9974

0.01

300 K

0.008 300.00

290 K

0.006 y = 0.0127x + 0.0032 R2 = 0.9984

0.004

200.00

0.002 100.00

0 0

0.00 290.00

310.00

330.00

350.00

370.00

390.00

410.00

λ (nm) Fig. 2. Effect of pioglitazone on fluorescence spectra of HSA (T = 300 K, pH 7.4, ex = 295 nm). CHSA : 1 × 10−6 mol L−1 ; Cpioglitazone × 10−7 mol L−1 : 0, 3.3, 6.6, 10.0, 20.0, 30.0 and 40.0, respectively.

The linearity of this plot proposes a single type of quenching process [11]. The linearity of Stern–Volmer plots for pioglitazone concentration lower than 1.0 × 10−6 mol L−1 indicate the existence of one or more binding sites equally accessible to the drug [8,16]. To interpret the data from fluorescence quenching studies, it is important to understand the nature of the interactions, which take place between the pioglitazone and serum albumins. As discussed earlier, the fluorescence quenching mechanism usually involves static or dynamic quenching. However, in this case the downward curvature toward x-axis at high pioglitazone concentration suggests the existence of static quenching [11]. 1.9

290 K

1.8

300 K

1.7 1.6

310 K

F0/F

1.5 1.4 1.3 1.2 1.1 1

0.05

0.1

0.15

0.2

1/[Q]x10-7

0.25

0.3

0.35

Fig. 4. The Lineweaver–Burk curves for quenching of HSA by pioglitazone in different temperature.

A linear Stern–Volmer plot is generally indicative of a single class of fluorophore, all equally accessible to quencher. If two fluorophore populations are present, and one class is not accessible to quencher, then the Stern–Volmer plots deviate from linearity toward the x-axis. This result is frequently found for the quenching of tryptophan fluorescence in proteins by polar or charged quenchers [13]. 3.4. Effect of temperature on fluorescence quenching of HSA Static and dynamic quenching can be distinguished by their deferring dependence on temperature. Higher temperatures result in faster diffusion and hence larger amount of collisional quenching. Higher temperatures will usually result in the dissociation of weekly bound complexes, and hence smaller amounts of static quenching [13]. The quenching of HSA with pioglitazone was also calculated according to the modified Stern–Volmer (Lineweaver–Burk equation) [17,18]: (F0 − F)−1 = F0−1 + Ka−1 F0−1 [Q ]−1

(4)

where Ka is the binding efficiency of micromolecules to biological macromolecules at ground state, which can be determined by the slope of the (F0 − F)−1 vs. [Q]−1 curves as shown in Fig. 4. It is exactly the opposite of the dynamic quenching definition, which can indicate that the quenching was not initiated by collision. Table 1 shows the quenching constants at different temperatures. As it can be seen from Table 1, quenching constants decrease with increasing temperature. These values are too large to be due to the collisional quenching. Less quenching in the higher temperature shows that complex formation is less stable at higher temperature. Table 1 The quenching constants for titration of HSA (1 × 10−6 mol L−1 ) by pioglitazone at different temperatures in pH 7.4.

0.9 0

2

4

6

8 10 12 14 16 18 20 22 24 26 28 30 32

[Q]x10 Fig. 3. The Stern–Volmer plots at em = 331 nm; CHSA : 1 × 10−6 mol L−1 .

7

different

temperatures.

ex = 295 nm;

Temperature (K)

KSV × 105 (L mol−1 )

R2

290 300 310

5.55 4.45 3.10

0.997 0.995 0.994

F. Faridbod et al. / Spectrochimica Acta Part A 78 (2011) 96–101

UV Spectra HSA-Pioglitazone

0.060000

14.2

0.030000

R2 = 0.9995

13.8

HSA-Pioglitazone

Ln Kb

Absorbtion

y = 5077.1x - 3.4468

14

0.050000

0.040000

99

HSA

13.6

13.4 0.020000

13.2 0.010000

Pioglitazone

13 0.000000 240

250

260

270

280

290

300

310

320

12.8

Wavelength (nm)

0.0032

0.0033

Fig. 5. Absorption spectra of pioglitazone, HSA, and HSA–pioglitazone complex; CHSA : (10−5 mol L−1 ); Cpioglitazone (10−5 mol L−1 ).

When small molecules bind independently to a set of equivalent sites on a macromolecule, the equilibrium between free and bound molecules is given by Eq. (5) [19,20]:

F − F  0 F

= log Kb + n log [Q ]

0.0034

0.0034

0.0035

0.0035

1/T (K) Fig. 6. Temperature dependence of binding constant at pH 7.4; HSA concentration: 1.0 × 10−6 mol L−1 .

3.5. Binding constant (Kb ) and number of binding sites (n)

log

0.0033

(5)

where Kb is the binding constant to a single site and n is the number of binding sites per HSA. According to Eq. (5), the binding constant Kb and the number of binding sites n can be obtained 1.1 × 105 mol L−1 and about 1 (n = 1.13), respectively. It can be concluded that pioglitazone molecule is binding HSA to form a 1:1 complex. This indicates that pioglitazone forms a reversible complex with HSA, i.e. the complex formation can be regarded to be a dynamic procedure. 3.6. UV–Vis study Fig. 5 shows the absorption spectrum of HSA, pioglitazone, and HSA–pioglitazone system. The absorption spectra of HSA–pioglitazone system (Fig. 5) are clearly different from those of HSA or pioglitazone alone, which is obvious evidence that they have formed at least one protein–drug complex which confirms the quenching of HSA with pioglitazone belongs to static quenching. And that is why the static binding constants were used when the thermodynamic parameters were calculated. 3.7. Determination of acting force Generally, the force between organic micromolecule and biological macromolecule may include hydrogen bond, van der Waal’s force, electrostatic force and hydrophobic: hydrophobic interactions force and so on [21].

Because the effect of the temperature is pretty small, the enthalpy change of the interaction can be regarded as a constant if the temperature range is not too wide. Therefore, from the following Eqs. (6)–(8): G◦ = −RT ln Ka

(6)

G◦ = H ◦ − T S ◦

(7)

ln Ka =

−H ◦ S ◦ + RT R

(8)

where Ka is the binding constant obtained from Eq. (4) at the corresponding temperature, and R is the gas constant. The temperatures used were 290, 300 and 310 K. The plot of ln Ka vs. 1/T (Fig. 6) allows the determination of H◦ and S◦ . The enthalpy change (H◦ ) is calculated from the slope of Eq. (6). Thus, to explain the interaction of pioglitazone with HSA, thermodynamic parameters were calculated. Table 2 shows the values of H◦ , G◦ and S◦ . The negative value of G◦ shows that the interaction process is spontaneous. van der Waal’s interactions and hydrogen bonds play major role in the protein ligand interaction [22]. The negative values of enthalpy (H) and the entropy (S) of the interaction of pioglitazone and HSA indicate that the binding is mainly enthalpy stabilized and the entropy destabilized. Desolvation of polar groups on HSA and hydrogen bonding are specific and directed, which may be the best identified through their negative enthalpy and entropy of complex formation. According to Förster’s non-radiation energy transfer theory, the energy transfer can occur only when the fluorescence emission spectra of the donor and the absorption spectra of the acceptor have enough overlap and the distance between donor and acceptor is not longer than 7 nm. In detail, it can be described by the

Table 2 The binding efficiency of micromolecules to biological macromolecules (Ka ) and thermodynamic parameters for pioglitazone–HSA complex. Temperature (K)

Ka × 105 (L mol−1 )

G◦ (kJ mol−1 )

H◦ (kJ mol−1 )

S◦ (J mol−1 K−1 )

R2

290 300 310

12.6 7.24 4.10

−33.902 −33.616 −33.329

−42.211 −42.211 −42.211

−28.65 −28.65 −28.65

0.998 0.997 0.998

100

F. Faridbod et al. / Spectrochimica Acta Part A 78 (2011) 96–101

Fig. 7. Overlap of the absorption spectrum of pioglitazone (Cpioglitazone = 10−5 mol L−1 ) and the fluorescence emission spectrum of HSA −5 −1 (CHSA = 10 mol L ).

following three equations [23,24]: E =1−

R6 F = 6 0 F0 R0 + r 6

R0 = 8.8 × 10−25 K 2 n−4 J



J=

F()ε(0



)4 

F()

(9) (10) (11)

where E is the efficiency of transfer between the donor and the acceptor, R0 the critical distance when the efficiency of transfer is 50%, r the distance between the acceptor and the donor, K2 the space factor of orientation, N the refracted index of medium,  the fluorescence quantum yield of the donor, J the effect of the spectral overlap between the emission spectrum of the donor and the absorption spectrum of the acceptor, F() the corrected fluorescence intensity of the donor in the wavelength range 0 to , and ε(0 ) the extinction coefficient of the acceptor at 0 . The overlap of the absorption spectrum of pioglitazone and the fluorescence emission spectrum of HSA is shown in Fig. 7. The overlap integral, J, can be evaluated by integrating the spectra in Fig. 7 according to Eq. (11). Under our experimental conditions, K2 = 2/3, n = 1.336 and  = 0.13, J can be evaluated to be 7.31 × 10−14 cm3 L mol−1 . Under these experimental conditions, R0 is 1.9 nm and energy transfer effect (E) is 0.34 from Eq. (9). Thus, the maximum distance between pioglitazone and amino acid residue in HSA, r is 2.1 nm. This confirms that the energy transfer between pioglitazone and HSA contributes to decrease the HSA fluorescence intensity. 3.8. Molecular docking study Experimental observations were followed up with docking studies where pioglitazone was docked to HSA to determine the preferred binding site on the protein. Three-dimensional structure of pioglitazone was constructed and optimized using Polak–Ribiere conjugate gradient algorithm and AMBER95 force field implemented in HyperChem (HyperCube Inc., Gainesville, FL). Then, in AutoDockTools package, the partial atomic charges were calculated using Gasteiger–Marsili method [25] and after merging non-polar hydrogens, rotatable bonds were assigned. Of more than 50 X-ray crystallographic structures related to human serum albumin, in Protein Data Bank [26], entry with PDB

Fig. 8. The schematic interaction of HSA and pioglitazon using molecular docking model.

ID:1BM0 [27] was chosen for dockings because of no missing atoms, no co-crystallized ligand and having a reasonably good resolution ˚ All the water molecules were removed Using a plain text (2.5 A). editor, then missing hydrogens and Kollman partial charges were added in AutoDockTools environment. Finally non-polar hydrogens were merged to their corresponding carbons, and desolvation parameters were assigned to each atom. Flexible-ligand docking studies were carried out using AutoDock version 3.0.5 [28]. In order to find potential binding sites of pioglitazone, the grids (one for each atom type in the ligand, plus one for electrostatic interactions) were chosen to be sufficiently large to include the whole HSA molecule. The points of the grids were thus 126 × 126 × 126 with a grid spacing of 0.375 A˚ (roughly a quarter of the length of a carbon–carbon single bond). Of the three different search algorithms offered by AutoDock, Lamarckian genetic algorithm (LGA) was applied for the local search, so-called Pseudo-Solis and Wets algorithm [29]. For all docking parameters, standard values were used as described before [28], except the amount of independent docking runs performed for each docking simulation which was set to maximum value supported by AutoDock (i.e. 256). Cluster analysis was performed on the docked results using a root mean square (RMS) tolerance of ˚ 0.5 A. The docking results are in good agreement with the experimental results and reveal important interactions between pioglitazone and HSA. The interaction between pioglitazone and HSA for the binding site is dominated by hydrogen binding interactions with Arg-257, Lys-274, His-440 and Leu-112 residues (Fig. 8). The distance between Trp-214 and the binding sites found by docking was 2.2 nm. The result is very close to that obtained by FRET calculation (2.1 nm). Considering HSA residues involving in the binding sites with pioglitazone, it became evident that the binding sites are located in the subdomain IIA and IIIA of HSA [27]. 4. Conclusion This paper presents spectroscopic and molecular docking studies on the interaction of pioglitazone with albumin protein using fluorescence emission and UV–Vis spectra. It was shown that the fluorescence of HSA has been quenched for reacting with pioglitazone. The quenching is a kind of static fluorescence quenching with non-radiation energy transfer happening within single molecule. The binding constant at 300 K was figured out to be 7.24 × 105 L mol−1 . The thermodynamic parameters agree with G◦ < 0, H◦ < 0, S◦ < 0. The binding locality is an area 2.1 nm

F. Faridbod et al. / Spectrochimica Acta Part A 78 (2011) 96–101

away from tryptophan residue-214 in HSA and the binding power between them is mainly hydrogen bonding interaction. References [1] Available online: http://en.wikipedia.org/wiki/Pioglitazone. [2] U. Smith, Int. J. Clin. Pract. Suppl. 121 (2001) 13–18. [3] L. Shargel, Applied Biopharmaceutics & Pharmacokinetics, McGraw-Hill, Medical Pub. Division, New York, 2005. [4] T. Peters, All About Albumin, Academic Press, San Diego, CA, 1996. [5] H.P. Rang, M.M. Dale, J. Ritter, Molecular Pharmacology, 3rd ed., Academic Press, New York, 1995. [6] M. Dockal, D.C. Carter, F. Rüker, J. Biol. Chem. 274 (1999) 29303–29310. [7] D.C. Carter, J.X. Ho, Advances in Protein Chemistry, vol. 45, Academic Press, San Diego, 1994. [8] W.E. Evans, J.J. Schentag, W.J. Jusko (Eds.), Applied Pharmacokinetics: Principles of Therapeutic Drug Monitoring, Applied Therapeutics, Vancouver, WA, 1992, pp. 5.1–5.48. [9] H.H. Parikh, K. McElwain, V. Balasubramanian, W. Leung, D. Wong, M.E. Morris, M. Ramanathan, Pharm. Res. 17 (2000) 632–637. [10] W. Bal, J. Christodoulou, P.J. Sadler, A. Tucker, J. Inorg. Biochem. 70 (1998) 33–39. [11] J.R. Lakowicz, Principles of Fluorescence Spectroscopy, 2nd ed., Kluwer Academic Publishers, Plenum Press, New York, 1999. [12] X.M. He, D.C. Carter, Nature 358 (1992) 209–215. [13] B. Zhou, Z.D. Qi, X. Qi, J.X. Dong, Y.Z. Zhang, Y. Liu, J. Biochem. Biophys. Methods 70 (2007) 743–747.

101

[14] A. Barzegar, A.A. Moosavi-Movahedi, N. Sattarahmady, M.A. Hosseinpour-Faizi, M. Aminbakhsh, F. Ahmad, A.A. Saboury, M.R. Ganjali, P. Norouzi, Protein Peptide Lett. 14 (2007) 13–18. [15] F.Q. Cheng, Y.P. Wang, Z.P. Li, C. Dong, Spectrochim. Acta A 65 (2006) 1144–1147. [16] L. Peng, H. Minbo, C. Fang, L. Xi, Z. Chaocan, Protein Peptide Lett. 15 (2008) 360–364. [17] S.S. Lehrer, Biochemistry 10 (1971) 3254–3263. [18] C.B. Murphy, Y. Zhang, T. Troxler, V. Ferry, J.J. Martin, W.E. Jones, J. Phys. Chem. B 108 (2004) 1537–1543. [19] H. Hou, Z. Qi, Y. OuYang, F. Liao, Y. Zhang, Y. Liu, J. Pharm. Biomed. Anal. 47 (2008) 134–141. [20] Z.F. Xi, C.L. Zhang, Talanta 47 (1998) 1223–1229. [21] D. Leckband, Measuring the forces that control protein interactions, Annu. Rev. Biophys. Biomol. Struct. 29 (2000) 1. [22] P.D. Ross, S. Subramanian, Biochemistry 20 (1981) 3096–3102. [23] L. Stryer, Science 162 (1968) 526–533. [24] W.M. Vaughan, G. Weber, Biochemistry 9 (1970) 464–473. [25] J. Gasteiger, M. Marsili, Tetrahedron 36 (1980) 3219–3228. [26] H.M. Berman, J. Westbrook, Z. Feng, G. Gilliland, T.N. Bhat, H. Weissig, Nucleic Acids Res. 28 (2000) 235–242. [27] S. Sugio, A. Kashima, S. Mochizuki, M. Noda, K. Kobayashi, Protein Eng. 12 (1999) 439–446. [28] G.M. Morris, D.S. Goodsell, R.S. Halliday, R. Huey, W.E. Hart, R.K. Belew, J. Comput. Chem. 19 (1998) 1639–1662. [29] F.J. Solis, R.J.B. Wets, Math. Oper. Res. 6 (1981) 19–30.