Study on interaction of mangiferin to insulin and glucagon in ternary system

Hui Lin, Rui Chen, Xiaoyan Liu, Fenling Sheng, Haixia Zhang *
State Key Laboratory of Applied Organic Chemistry, Lanzhou University, Lanzhou 730000, China
Corresponding author: Haixia Zhang Tel: +86-931-4165997 Fax: +86-931-8912582 E-mail: zhanghx@lzu.edu.cn

Abstract:

The binding of mangiferin to insulin and glucagon was investigated in the presence and absence of another

Peptide by optical spectroscopy. Fluorescence titration experiments revealed that mangiferin quenched the

intrinsic fluorescence of insulin and glucagon by static quenching. The ratios of binding constants of glucagon-

mangiferin to insulin-mangiferin at different temperatures were calculated in pure and ternary system,

respectively. The results indicated that the Peptides were competitive with each other to act on mangiferin.

Values of the thermodynamic parameters and the experiments of pH effect proved that the key interacting forces

between mangiferin and the Peptides were hydrophobic interaction. In addition, UV-Vis absorption, synchronous

fluorescence and Fourier transform infrared measurements showed that the conformation of insulin and glucagon

were changed after adding mangiferin.

Keywords: Mangiferin, Insulin, Glucagon, Spectroscopy, Binding parameters

1. Introduction

A lot of studies related with interaction between small drugs and transport proteins have been

widely investigated in the past few years [1-4]. The binding between small molecules and proteins

can elucidate the properties of small moleculeprotein complex, provide useful information of the

delivery or structural features of small molecules, and influence on the concentration, distribution

1
and conformation of target protein in body synchronously [5]. However, few studies were related

with the interaction of small drugs and active Peptides in ternary system. In blood, there are a lot of

active Peptides which form transient biological interactions with other substances including drugs,

protein and so on [6]. These transient biological interactions are dynamic and control the local

concentrations of interacting biological pair. Owing to the low concentrations of the Peptides, the

transient biological interactions can affect not only the concentrations and forms of Peptides but

also the delivery of foreign compounds such as drugs. Thus, it is significant equally to study the

interaction of small drugs and Peptides.

Diabetes mellitus is found increasing rapidly in all the world. People suffering from diabetes

are not able to produce or properly use insulin in the body and have a high content of blood glucose.

Both of insulin and glucagon are important Peptides in blood glucose regulation and diabetes

treatment. Insulin plays important roles in lowering blood glucose, regulating lipid metabolism,

inhibiting of protein degradation, promoting protein synthesis, and affecting growth [7]. Glucagon

possess significant roles in increasing blood glucose, promoting ketogenesis, regulating lipid

metabolism, and affecting protein metabolism [8]. Both Peptides have the adversed functions and

their concentrations should keep the relatively stable. Up to now, there is no report related with the

interaction between small drug molecule and the two Peptides.

Most papers related with interaction of small molecule and protein investigated that in a pure

system [9-12], in which only the studied protein existed in the solution. Here, we firstly try to study

the interaction of small molecule and Peptide under the coexistance of another Peptide. Mangiferin

(MA) was chosen as the model small molecule, which is the active component of Chinese medicine

Anemarrhena asphodeloides Bge., and has a lot of beneficial functions including curing Diabetes.

2
Its binding to Bovine serum albumin (BSA) and Human serum albumin (HSA) had been studied

[13-14]. Spectral analysis was chosen here as a tool to study the interaction because of the

convenient manipulation, low spending and abundant theory foundation [15].

2. Materials and methods

2.1 Materials

Insulin (from bovine pancreas) was obtained from Roche Diagnostics Co. (Indianapolis, U.S.A).

Glucagon was purchased from Chengdu Kaijie Bio-pharmaceutical Co. Ltd. (Chengdu, China). Stock

solution of insulin and glucagon were prepared in buffer solution and kept in the dark at 277K. MA

(90%) was obtained from Chengdu tianfutang Economy Co. Ltd. (Chengdu, China). Stock solution of

MA (810-5 mol L-1) was prepared in buffer solution for FT-IR experiments and stock solution of MA

(1.010-3 mol L-1) was prepared in ethanol for UV-vis and fluorescence experiments. The buffer

solution (pH 7.40) consisted of Tris (0.1 mol L-1) and HCl (0.1 mol L-1). pH was checked with a

PHSJ-3F pH-meter (Shanghai Precision Scientific Instruments Co. Ltd, Shanghai, China). Other

chemicals were of analytical grade and the double-distilled water was used throughout the

experiments.

2.2 Apparatus and methods

The UV-vis absorption spectra were recorded on a TU-1810 spectrophotometer (Purkinje,

China) at 298 K in the range of 250-400 nm equipped with 1.0 cm path length quartz cells.

Fluorescence measurements were recorded on RF-5301PC fluorophotometer (Shimadzu,

Japan) using 5nm/5nm bandwidths at excitation and emission wavelength of 276 nm and 300-500

nm, respectively. Synchronous fluorescence spectra were recorded with the excitation and emission

slit widths of 5/5nm, and the range of synchronous scanning were: = 15 nm, = 60 nm.

Fluorescence titration experiment was accomplished as following: 3.0 mL Peptides (insulin,

3
glucagon or mixture of them) solution at concentration of 1.5 10-6 mol L-1, was titrated by

successive additions of 10 L stock solution of MA (1.010-3 mol L-1). Titrations were done

manually by using micro-injector, and the fluorescence spectra were then measured at three

temperatures (296, 303, and 310 K).

FT-IR measurements were carried out at room temperature on a Nicolet Nexus 670 FT-IR

spectrometer (USA) equipped with a Germanium attenuated total reflection (ATR) accessory, a

DTGS KBr detector and a KBr beam splitter. All spectra were taken via the Attenuated Total

Reflection (ATR) method with a resolution of 4 cm-1 and 60 scans. Spectra processing procedures:

The absorbance of reference solution from the spectra of sample solution was subtracted to get the

FT-IR spectra of proteins. The subtraction criterion was that the original spectrum of protein solution

between 2200 and 1800 cm-1 was featureless [16].

3. Results and discussion

3.1 Effect of MA on Peptides fluorescence spectra

The fluorescence spectra of insulin, glucagon and the mixture of insulin and glucagon in the

presence of different concentrations of MA were shown in Fig. 1 (a), Fig. 1 (b) and, Fig. 1 (c),

respectively. Insulin shows a fluorescence emission with a peak at 307 nm and the excitation

wavelength of 276 nm, and MA is almost none fluorescent under the present experiment conditions

[13]. With the increasing concentration of MA, the fluorescence intensity of insulin decreased

remarkably. However, the maximum emission wavelength was hardly changed. The results

indicated the formation of a complex between MA and Insulin.

Glucagon shows a strong fluorescence emission with a peak at 346 nm at the excitation

4
wavelength of 276 nm. With the increasing concentration of MA, the fluorescence intensities of

glucagon decreased remarkably too, and there was obvious em blue shift of maximum emission

wavelength with the addition of MA. It implied formation of a complex between MA and glucagon

and changes in Peptide conformation. The fluorescence intensity of the mixture of insulin and

glucagon decreased at 307nm and 344nm, respectively, which indicated the information of both

Peptide.. The quenching phenomena were similar to those in Fig. 1(a) and Fig. 1(b).

(Fig. 1 near here)

3.2 Quenching mechanism.

A variety of molecular interactions can result in quenching, including excited-state reactions,

molecular rearrangements, energy transfer, ground-state complex formation, and collisional

quenching [17]. Fluorescence quenching can proceed via different mechanisms usually classified as

dynamic quenching, static quenching or simultaneous static and dynamic quenching. For dynamic

quenching, fluorescence quenching can be described by Stern-Volmer equation [18]:

F0/F = 1 + kq0 [Q] = 1 + Ksv [Q] (1)

F0 and F are fluorescence intensities of biomolecule before and after addition of quencher,

respectively. kq is the quenching rate constant of the biomolecule (kq : L mol-1 s-1), Ksv is the Stern

Volmer dynamic quenching constant (Ksv : L mol-1), 0 is the average lifetime of the biomolecule

without quenching (0: 10-8 s) and [Q] is the concentration of quencher. In order to investigate the

quenching mechanism, the procedure of the fluorescence quenching was first assumed to be a

dynamic quenching progress. SternVolmer plots of insulin and glucagon with MA were shown in

Fig. 2. The corresponding SternVolmer dynamic quenching constants at different temperatures

5
were summarized in Table 1. It can be found the SternVolmer dynamic quenching constants of

insulin-MA in pure and ternary system were almost same within error limit, which indicated that

the quenching rate was independent of temperature and was not influenced by the existence of

glucagon. The same conclusion was obtained for the system of glucagon and MA. For dynamic

quenching, the maximum scatter collision quenching constant kq of various quenchers with the

biopolymer is 21010 L mol1 S1 [19]. The values of kq in all systems studied at three temperatures

were much greater than it, which indicated that the quenching of insulin and glucogon are not

initiated by dynamic quenching, but probably by static quenching resulting from the formation of

insulin-MA complex or glucagon-MA complex [20].

Comparing with the Ksv of the complex of MA-BSA [13], to those of insulin-MA and

glucagon-MA, it was found MA could quench the fluorescence of BSA much more effectively. For

investigating the relationship between Ksv and the molecular volumes of Peptides, the molecular

volumes were calculated roughly with hyperchem software (Hyperchem 7.0). The results showed

that the Ksv has no relationship with Peptide volumes (insulin: 10805.69 3, glucogon:7396.42 3 ).

(Fig.2 and table 1 near here)

3.3 Analysis of binding constants and binding sites

For static quenching, the relationship between fluorescence quenching intensity and the

concentration of quenchers can be described by Eq. (2) [21, 22]:

lg [(F0-F)/F] = lgKa + nlgQ (2)

where Ka is the binding constant (Ka: L mol-1) and n is the number of binding sites per the

biomolecule with fluorophore.

Fig. 3 shows the double-logarithm curves of insulin-MA (a) and glucagon-MA (b) in pure

and ternary system under different temperatures. The results of the all systems studied were

6
summarized in Table 2. All correlation coefficients were larger than 0.995, which indicated that the

interaction between Peptide (insulin or glucagon) and MA agreed well with the site-binding model

underlied in Eq. (2).

From Fig. 4, it could be found the ratios of binding constants of glucagon-MA to insulin-MA at

303 K and 310 K in pure system were higher than those in ternary system, but lower a little at 296

K. Glucagon acted strongerly with MA than insulin in pure system under 310 K, which is closed

to the normal body temperature. In ternary system, the interaction of glucagon-MA became much

weaker and the interaction of insulin-MA became much stronger until both had similar binding

constants. The solution of insulin is a dynamic equilibrium of monomers, dimers, tetramers, and so

on, depending on the concentration, pH, ionic strength, temperature, and solvent composition [23-

25]. The insulin may exist as more dimers on the condition of pure system than in ternary system.

The binding site was possibly hidden in the core of dimers, which hindered the binding with small

molecule MA. In the ternary system, glucagon may destroyed some insulin dimers and some

binding sites of insulin were exposed, which increased the possibility of binding to MA.

Meanwhile, owing to glucagon having smaller molecular size, which was maybe enwrapped by

insulin and lost the competition to bind MA.

According to Eq. (2), the binding sites of the Insulin-MA and Glucagon-MA were calculated to

be 1.00 approximately. The binding sites in pure system were also measured by UV-vis

experiments with mole ratio method [26]. It was found the measured results were consistent with

the calculated above.

(Fig 3, Fig 4, Table 2 near here)

3.4 Thermodynamic parameters and nature of the binding forces

7
 The interaction forces between drug and biomolecule may involve hydrophobic force,

electrostatic interactions, van der Waals interactions, hydrogen bonds, steric contacts within the

antibody-binding site, etc. In order to elucidate the interaction of MA to insulin and glucagon, the

thermodynamic parameters were calculated from Eqs. (3)-(4). If the temperatures does not vary

significantly, the enthalpy change (H) can be regarded as a constant and were calculated

according to the vant Hoff equation:

ln k = -H/RT + S/R (3)

Thus, a plot of ln k versus 1/T enabled the determination of H and S for the binding. The free

energy change (G) was estimated from the following equation:

G = H -TS (4)

Where k is the binding constant at the corresponding temperature and R is the gas constant. The

thermodynamic parameters for the interaction of MA to insulin and glucagon were summarized in

Table 2. The negative sign for G means that the interaction progress was spontaneous. The

positive H and S values indicated that hydrophobic force may play a major role in the binding

between MA and the both Peptides [27].

The effect of pH values on the binding constants of Insulin -MA and Glucagon-MA was studied

(Fig 5). The trend of change on Ka in ternary system was similar to that in pure system. The

values of binding constants of Peptides-MA are biggest at pH 7.40 in pure and ternary systems.

MA existed as ionic form and was negatively charged while pH 7.40 [28]. The isoelectric points of

insulin and glucagons are 5.35 and 7.00, respectively [29], the charges of Peptides surface are

negative at pH 7.40. Hence, electrostatic attraction force and hydrogen bond interaction were not

the key roles at this environment, and hydrophobic force may play a major role in the binding

8
between MA and Peptides (insulin or glucagon).

(Fig. 5 near here)

3.5 Conformation investigation

To explore the structural change of insulin and glucagon, UV-vis absorption, synchronous

fluorescence and FT-IR were performed.

The UV-vis spectrum of MA, insulin, glucagon, insulin-MA and glucagon-MA were recorded

as shown in Fig. 6. With the increasing concentration of MA, the UV intensities of insulin and

glucagon increased at 280 nm and had a blue shift due to formation of complex between MA to

proteins [14, 30]. The results implied that the microenvironment around chromophore of insulin and

glucagon was changed upon addition of MA.

(Fig .6 near here)

The synchronous fluorescence spectra offers the characteristics of tyrosine residues and the

tryptophan residue of protein when the wavelength interval () is 15 nm and 60nm, respectively

( = emission - excitation) [31]. Insulin has no tryptophan residues, synchronous fluorescence spectra

of insulin upon addition of MA gained at = 15 nm are shown in Fig. 7 (a) and (c). It can be found

that the fluorescence intensity of tyrosine decreased and had a slightly red shift at maximum

emission in pure system, which indicated that the tyrosine residues are placed in a more

hydrophilic environment and their micro-environment is rearranged [31] thus resulted in the

conformational changes of insulin. However, the fluorescence intensity of tyrosine had no shift in

ternary system, the possible reason was that the synchronous fluorescence spectra of glucagon

influenced the spectra of insulin. Glucagon has both of tyrosine residues and tryptophan residues.

However, tryptophan residues have stronger fluorescence intensities. Fig. 7 (b) and (d) gave the

synchronous fluorescence spectra of glucagon upon addition of MA gained at = 60 nm, which

9
shows that the fluorescence intensity of tryptophan decreased and had a red shift at maximum

emission in pure and ternary systems. It implied the conformation of glucagon is changed due to

that tryptophan residues are exposed more to the aqueous phase [31].

(Fig. 7 near here)

Detailed information of insulin and glucagon conformation after adding MA was investigated

with FT-IR difference spectra as shown in Fig. 8. The spectra of Fig. 8 a (1) and Fig. 8 b (1) were

spectra of insulin before and after adding MA, respectively. The spectra of Fig. 8 a (2) and Fig. 8 b

(2) were spectra of glucagon before and after adding MA, respectively. From Fig. 8, it concluded

that the secondary structure of insulin and glucagon were changed after the addition of MA, because

the peak position of amide I band (insulin: 1643.28 cm-1, glucagon: 1651.65 cm-1) and amide II band

(insulin: 1559.60 cm-1, glucagon: 1546.85 cm-1) had small shifts and their peak shapes were also

slightly changed [32-35]. The results confirmed that there was a reaction happened between MA

and the both Peptides.

(Fig. 8 near here)

A quantitative analysis of the protein secondary structure of insulin and glucagon before and

after the interaction with MA in Tris-HCl buffer was finished. Based on the literature [36], the

component bands of amide I was attributed according to the well-established assignment criterion.

Then the percentages of each secondary structure of insulin and glucagon can be calculated and

Table 3 shows the calculated results. It was found that the -turn, -helical, random coil and -sheet

structures in the secondary structure of insulin and glucagon were changed after adding MA. The

results confirmed that the binding of MA to insulin and glucagon had changed the secondary

structure of insulin and glucagon.

10
(Table 3 near here)

4. Conclusion

In this paper, the binding MA to insulin and glucagon has been investigated for the first time

by fluorescence method combined with UV-vis and FT-IR spectroscopy techniques under

simulative physiological condition. The results suggested that MA quenches the intrinsic

fluorescence of insulin and glucagon through static quenching mode. Conformations of both

Peptides were changed slightly after adding MA. Hydrophobic interactions played a key role in the

binding progress of MA to insulin and glucagon. The experimental results indicated that insulin

outvied glucagons to bind MA at 303 K and 310 K when they are coexisted in the same system.

Acknowledgement

This work was supported by the National Natural Science Foundation of China (NSFC) Fund

(No.20775029), Program for New Century Excellent Talents in University (NCET-07-0400) and the

Central Teacher Plan in Lanzhou University.

References

[1] F.-Q. Cheng, Y.-P. Wang, Z.-P. Li, C. Dong, Spectrochim. Acta A 65 (2006) 11441147

11
[2] J.Q. Liu, J.N.Tian, Z.D. Hu, X.G. Chen, Bioorg. Med.Chem.73 (2004) 443450.

[3] S.Y. Bi, D.Q. Song, Y.H. Kan, D. Xu, Y. Tian, X. Zhou, H.Q. Zhang, Spectrochim. Acta A 62

(2005) 203212

[4] Y.H. Zhang, L.J. Dong, J.Z. Li, X.G. Chen, Talanta. 76 (2008) 246-253.

[5] D.H. Ran, X. Wu, J.H. Zheng, J.H. Yang, H.P. Zhou, M.F. Zhang, Y.J. Tang, J. Fluoresc. 17

(2007) 721-726.

[6] S. Ohlson, S. Shoravi, T. Fex, R. Isaksson, Anal. Biochem. 359 (2006) 120-123

[7] M.Y. Xu, G.H. Lu, M.D. Chen, Diabetes Mellitus, First ed., Shanghai Scientific and Technical

Publishers, Shanghai, 2003, pp 55-56.

[8] M.Y. Xu, G.H. Lu, M.D. Chen, Diabetes Mellitus, First ed., Shanghai Scientific and Technical

Publishers, Shanghai, 2003, pp 60-61.

[9] M. Bogdan, A. Pirnau, C. Floare,C. Bugeac, J. Pharm. Biomed. Anal. 47 (2008) 981-984.

[10] Y.-J. Hu, W. Li, Y. Liu, J.-X. Dong, S.-S. Qu, J. Pharm. Biomed. Anal. 39 (2005) 740-745.

[11] X.-H. Liu, P.-X. Xi, F.-J. Chen, Z.-H. Xu, Z.-Z. Zeng, J. Photochem. Photobio. B. 92 (2008)

98-102.

[12] P.B. Kandagal, S.S. Kalanur,D.H. Manjunatha, J.Seetharamappa, J. Pharm. Biomed. Anal. 47

(2008) 260267.

[13] H. Lin, J. F. Lan, M. Guan, F. L. Sheng, H. X. Zhang, Spectrochim. Acta A 73 (2009) 936941.

[14] Y.Y. Yue, X.G. Chen, J. Qin, X.J. Yao, J. Pharm. Biomed. Anal. 49 (2009) 753-759.

[15] J.H. Li, C.L. Ren, Y.H. Zhang, X.Y. Liu,X.J. Yao, Z.D. Hu, J. Mol. Struct. 885 (2008) 6469.

[16] T.Yuan, A.M. Weljie, H.J.Vogel, Biochem. 37 (1998) 31873195.

[17] F.L. Cui, L.X. Qin, G.S. Zhang, Q.F. Liu, X.J. Yao, B.L. Lei, J. Pharm. Biomed. Anal. 48

(2008) 1029-1036.

12
[18] J.R. Lakowicz, Principles of Fluorescence Spectroscopy, second ed., Plenum Press, New York,

1999, 237-265.

[19] W.R. Ware, J. Phys. Chem. 66 (1962) 455-458.

[20] Y.-P. Wang, Y.-L. Wei, C. Dong, J. Photochem. Photobiol. A 177 (2006) 6-11.

[21] X.Z. Feng, R.X.Jin, Y.Qu, X.W. He, Chin.Univ.17 (1996) 866.

[22] X.F. Wei, H.Z. Liu, J. Chin. Aanl. Chem. 28 (2000) 699.

[23] Y. Zhang, Y.J. Deng, M. Lu, J. Colloid Interf. Sci 337 (2009) 322331.

[24] W.D. Lougheed, H. Woulfe-Flanagan, J.R. Clement, A.M. Albisser, Diabetologia 19 (1980) 1.

[25] J. Brange, L. Langkjoer, Pharm. Biotechnol. 5 (1993) 315.

[26] C. D. Chriswell, A. A. Schilt, Anal. Chem. 47 (1975) 1623-1629

[27] P.D. Ross, S. Subramanian, Biochem. 20 (1981) 3096-3102.

[28] B. Gmez-Zaleta, M. T. Ramrez-Silva, A. Gutirrez, E. Gonzlez-Vergara, M. Gizado-Rodr

guez, A. Rojas-Hernndez, Spectrochim. Acta A 64 (2006) 1002-1009.

[29] A.S.R. Andrade, L. Vilela, H. Tunes, Braz. J. Med. Biol. Res. 30 (1997) 1421-1426.

[30] K.H. Ulrich, Pharmacol. Rev. 33 (1981) 1753.

[31] C-X. Wang, F-F. Yan, Y-X. Zhang, L. Ye, J. Photochem. Photobiol. A. 192 (2007) 23-28.

[32] S. Wi, P. Pancoka, T. A. Keiderling, Biospectroscopy. 4 (1998) 93-106.

[33] K. Rahmelow, W. Hubner, Anal. Biochem. 241 (1996) 5-13.

[34] X.-H. Liu, P.-X. Xi, F.J. Chen, Z.H. Xu, Z.Z. Zeng, J. Photochem. Photobiol. B. 92 (2008) 98-

102.

[35] Y. Li, W.Y. He, Y.M. Dong, F.L. Sheng, Z.D. Hu, Bioorgan. Med. Chem. 14 (2006) 1431-1436.

[36] M. Jiang, M.X. Xie, D. Zheng, Y. Liu, X.Y. Li, X. Chen, J. Mol. Struct. 692 (2004) 71-80.

13
Captions

Fig. 1. Fluoresence quenching spectra of insulin (a), glucagon (b) and the mixture of them (c) at

different concentrations of MA; Cprotein = 1.5 M; CMA from a to g: 0 M (a); 3.33 M (b); 6.66 M

(c); 9.99 M (d); 13.32 M (e); 16.65 M (f); 19.98 M (g). ex = 276 nm.

Fig. 2. Stern-Volmer plots for quenching of insulin and glucagon by MA in pure and ternary

system at 296, 303 and 310 K.

14
Fig. 3. Plots of log [F0-F)/F] vs. log [Q] in pure and ternary system at 296, 303 and 310 K.

Fig. 4. Plots of Ka1(Glucagon-MA)/Ka2(Insulin-MA) vs. T.

Fig. 5. Plots of Ka vs. pH in pure and ternary system.

Fig. 6. UV absorption spectra of insulin-MA (a) and glucagon-MA (b): C protein = 15 M; CMA from b

to g: 0M (b); 3.33 M (c); 6.66 M (d); 9.99 M (e); 13.32 M (f); 16.65 M (g); the spectra of

MA, CMA = 3.33 M (a).

Fig. 7. Effect of MA on the synchronous spectra of insulin and glucagon in pure and ternary

system. Cprotein = 15 M, CMA from a to g: 0 M (a); 3.33 M (b); 6.66 M (c); 9.99 M (d); 13.32

M (e); 16.65 M (f); 19.98 M (g).

=15 nm (a, c) (insulin, ex = 240 nm); =60 nm (b, d) (glucagon, ex = 240 nm).

Fig. 8. FT-IR spectra of free insulin (a1) and glucagon (b1). Difference spectra [(protein solution +

MA solution) (MA solution) in Tris-HCl buffer (pH 7.40), insulin (a2), glucagon (b2). (Cprotein = 10

M CMA = 30 M).

15