milk b-lactoglobulin complexes with tea polyphenols
TRANSCRIPT
Food Chemistry 127 (2011) 1046–1055
Contents lists available at ScienceDirect
Food Chemistry
journal homepage: www.elsevier .com/locate / foodchem
Milk b-lactoglobulin complexes with tea polyphenols
C.D. Kanakis a, Imed Hasni b, Philippe Bourassa b, P.A. Tarantilis a, M.G. Polissiou a,⇑,Heidar-Ali Tajmir-Riahi b,⇑a Laboratory of Chemistry, Department of Science, Agricultural University of Athens, 75 Iera Odos, 118 55 Athens, Greeceb Département de Chimie-Biologie, Université du Québec à Trois-Rivières, C.P. 500, Trois-Rivières, Québec, Canada G9A 5H7
a r t i c l e i n f o
Article history:Received 30 July 2010Received in revised form 1 December 2010Accepted 20 January 2011Available online 27 January 2011
Keywords:MilkTeaPolyphenolb-LactoglobulinSecondary structureFTIRCDFluorescence spectroscopyModelling
0308-8146/$ - see front matter � 2011 Elsevier Ltd. Adoi:10.1016/j.foodchem.2011.01.079
Abbreviations: b-LG, beta-lactoglobulin; C, cateepicatechin gallate; EGCG, epigallocatechin gallate; Fred; CD, circular dichroism.⇑ Corresponding authors. Tel.: +30 210 529 4241;
Polissiou); tel.: +1 819 376 5011x3310; fax: +1 819 3E-mail addresses: [email protected] (M.G. Polissiou), h
(H.-A. Tajmir-Riahi).
a b s t r a c t
The effect of milk on the antioxidant capacity of tea polyphenols is not fully understood. The complexa-tion of tea polyphenols with milk proteins can alter the antioxidant activity of tea compounds and theprotein secondary structure. This study was designed to examine the interaction of b-lactogolobulin(b-LG) with tea polyphenols (+)-catechin (C), (�)-epicatechin (EC), (�)-epicatechin gallate (ECG) and(�)-epigallocatechin gallate (EGCG) at molecular level, using FTIR, CD and fluorescence spectroscopicmethods as well as molecular modelling. The polyphenol binding mode, the binding constant and theeffects of polyphenol complexation on b-LG stability and secondary structure were determined. Struc-tural analysis showed that polyphenols bind b-LG via both hydrophilic and hydrophobic interactions withoverall binding constants of KC–b-LG = 2.2 (±0.8) � 103 M�1, KEC–b-LG = 3.2 (±1) � 103 M�1, KECG–b-LG = 1.1(±0.6) � 104 M�1 and KEGCG–b-LG = 1.3 (±0.8) � 104 M�1. The number of polyphenols bound per proteinmolecule (n) was 1.1 (C), 0.9 (EC), 0.9 (ECG) and 1.3 (EGCG). Molecular modelling showed the participa-tion of several amino acid residues in polyphenol–protein complexation with extended H-bonding net-work. The b-LG conformation was altered in the presence of polyphenols with an increase in b-sheetand a-helix suggesting protein structural stabilisation. These data can be used to explain the mechanismby which the antioxidant activity of tea compounds is affected by the addition of milk.
� 2011 Elsevier Ltd. All rights reserved.
1. Introduction
b-Lactoglobulin (b-LG) (Scheme 1), is the most abundantprotein of milk whey (1 g/l) and of a major interest in the foodindustry because of its nutritional and functional properties. Thestructure of this protein is well known (McKenzie & Sawyer,1967). At neutral pH, b-LG exists as a mixture of monomers and di-mers of which the equilibrium ratio depends on the associationconstant of the dimer and on the protein concentration. Eachmonomer consists of 162 amino acid residues and has a molecularmass of 18 kDa (Qin et al., 1998). As a member of the lipocalycinfamily, b-LG is a small globular protein folded into a calyx formedby eight antiparallel b-strands and an a-helix located at the outersurface of the b-barrel. b-LG exhibits strong affinity for a variety
ll rights reserved.
chin; EC, epicatechin; ECG,TIR, Fourier transform infra-
fax: +30 210 529 4265 (M.G.76 5084 (H.-A. Tajmir-Riahi)[email protected]
of hydrophobic and amphiphilic compounds, including fatty acids,phospholipids and aromatic compounds (Liang & Subirade, 2010).
Tea is one of the most popular beverages in the world. In recentyears, much has been said about the added health benefits or re-duced benefits of adding milk to your cup of tea (Stanner, 2007).The effect of milk on the antioxidant activity of tea polyphenolshas been the subject of much controversy. The dual effects of milkon the antioxidant capacity of different tea polyphenols were re-cently reported (Dubeau, Samson, & Tajmir-Riahi, 2010). However,the mechanism by which the antioxidant activity of tea com-pounds affected by milk is not yet known. In a recent report, thecomplexation of tea polyphenols with milk a- and b-caseins wasinvestigated and the effect of such interaction on the protein sec-ondary structure and the antioxidant capacity of tea polyphenolswas determined (Hasni et al., 2011). Plant polyphenolic com-pounds show strong interaction with globular proteins and cancause protein unfolding. In solution polyphenols such as catechins(Scheme 2) can form insoluble complexes with milk proteins(Liang & Xu, 2003). The binding affinity of polyphenols to proteinis size dependent and increases with their molecular size (De Fre-tias & Mateus, 2001). Larger polyphenols like those in black tea aremost likely to form complexes with milk proteins. This binding canaffect the electron donation capacity of the catechins by reducing
Scheme 1. 3D structure of b-lactoglobulin.
C.D. Kanakis et al. / Food Chemistry 127 (2011) 1046–1055 1047
the number of hydroxyl groups available in the solution. Studies inthe past showed the effects of milk protein on the antioxidantcapacity of tea polyphenols, whilst they did not address the effectof polyphenol complexation on the stability and conformation ofmilk proteins (Aguie-Beghin, Sausse, Meudec, Cheynier, & Doul-lard, 2008; Alexandropoulou, Komaitis, & Kapsokefalou, 2006; Job-stl, Howse, Fairclough, & Williamson, 2006; Kartsova & Alekseeva,2008; Shukla, Narayanan, & Zanchi, 2009; Yan, Hu, & Yao, 2009).Therefore, the structural characterisation of the interaction be-tween milk proteins and polyphenols is a major step in elucidatingthe induced effect of polyphenols on milk protein structure and onthe antioxidant activity of tea compounds.
Fluorescence quenching is considered as a technique for mea-suring binding affinities. Fluorescence quenching is the decreaseof the quantum yield of fluorescence from a fluorophore inducedby a variety of molecular interactions with quencher molecule(Lakowicz, 1999; Tayeh, Rungassamy, & Albani, 2009). Therefore,
Scheme 2. Chemical structu
it is possible to use quenching of the intrinsic tryptophan fluores-cence of Trp-61 and Trp-19 in b-LG (Liang & Subirade, 2010) as atool to study the interaction of polyphenol with b-LG in an attemptto characterise the nature of tea catechin–protein complexation.
We now present spectroscopic analysis and docking studies ofthe interaction of b-LG with several tea polyphenols catechin, epi-catechin, epicatechin gallate and epigallocatechin gallate (Scheme2) in aqueous solution at physiological conditions, using constantprotein concentration and various polyphenol contents. Structuralanalysis regarding catechin binding mode and the effects of cate-chin–protein complexation on the b-LG stability and secondarystructure is reported here.
2. Materials and methods
2.1. Materials
b-Lactoglobulin (A variant, purity > 90%) and tea catechins werepurchased from Sigma–Aldrich Chemical Co. (St. Louis, MO) andused as supplied. Other chemicals were of reagent grade and usedwithout further purification.
2.2. Preparation of stock solutions
b-Lactoglobulin was dissolved in aqueous solution (9 mg/ml toobtain 0.5 mM protein content) containing 10 mM Tris–HCl buffer(pH 7.4). Polyphenol 2 mM was prepared in Tris–HCl and diluted tovarious concentrations in Tris–HCl (1, 0.5 and 0.25 mM). The pro-tein concentration was determined spectrophotometrically usingthe extinction coefficients of 17,600 M�1 cm�1 (MW = 18 kD) at280 nm (Collini, Alfonso, & Baldini, 2000; Liang & Subirade, 2010).
2.3. FTIR spectroscopic measurements
Infrared spectra were recorded on a FTIR spectrometer (Nicolet6700), equipped with deuterated triglycine sulphate (DTGS) detec-
res of tea polyphenols.
Fig. 1. FTIR spectra in the region of 1800–600 cm�1 of hydrated films (pH 7.4) for free b-lactoglobulin (0.25 mM), free C (A) (0.5 mM), free EC (B) (0.5 mM), free ECG (C)(0.5 mM) and free EGCG (D) (0.5 mM) with difference spectra (diff.) of b-LG-polyphenol complexes (bottom two curves) obtained at different polyphenol concentrations(indicated on the figure).
1048 C.D. Kanakis et al. / Food Chemistry 127 (2011) 1046–1055
tor and KBr beam splitter, using ZnSe windows. A solution of pol-yphenol was added dropwise to the protein solution with constant
stirring to ensure the formation of homogeneous solution and toreach the target polyphenol concentrations of 0.125, 0.25 and
Fig. 2. Second derivative resolution enhancement and curve-fitted amide I region(1720–1575 cm�1) for free b-lactoglobulin and for b-lactoglobulin-polyphenolcomplexes (0.5 mM polyphenol and 0.25 mM protein concentrations at pH 7.4).
C.D. Kanakis et al. / Food Chemistry 127 (2011) 1046–1055 1049
0.5 mM with a final protein concentration of 0.25 mM. Spectrawere collected after 2 h incubation of b-lactoglobulin withpolyphenol solution at room temperature, using hydrated films.Interferograms were accumulated over the spectral range 4000–600 cm�1 with a nominal resolution of 2 cm�1 and 100 scans. Thedifference spectra [(protein solution + polyphenol solution) � (pro-tein solution)] were generated using a water combination modearound 2300 cm�1, as standard (Dousseau, Therrien, & Pezolet,1989). When producing difference spectra, this band was adjustedto the baseline level, in order to normalise difference spectra.
2.4. Analysis of protein conformation
Analysis of the secondary structure of b-lactoglobulin and itspolyphenol complexes was carried out on the basis of the proce-dure already reported (Byler & Susi, 1986). The protein secondarystructure is determined from the shape of the amide I band, locatedat 1660–1650 cm�1. The FTIR spectra were smoothed, and theirbaselines were corrected automatically using the built-in softwareof the spectrophotometer (OMNIC ver. 7.3). Thus the root-meansquare (rms) noise of every spectrum was calculated by means ofthe second derivative in the spectral region 1720–1575 cm�1. Fivemajor peaks for b-lactoglobulin and the complexes were resolved.The above spectral region was deconvoluted by the curve-fittingmethod with the Levenberg-Marquadt algorithm and the peakscorresponding to a-helix (1657–1651 cm�1), b-sheet (1634–1608 cm�1), turn (1670–1667 cm�1), and b-antiparallel (1691–1686 cm�1) were adjusted and the area was measured with theGaussian function. The area of all the component bands assignedto a given conformation were then summed up and divided bythe total area. The curve fitting analysis was performed using theGRAMS/AI Version 7.01 software of the Galactic IndustriesCorporation.
2.5. Circular dichroism
CD spectra of b-lactoglobulin and its polyphenol complexeswere recorded with a Jasco J-720 spectropolarimeter. For measure-ments in the far-UV region (178–260 nm), a quartz cell with a pathlength of 0.01 cm was used in nitrogen atmosphere. Protein con-centration was kept constant (12.5 lM), whilst varying each poly-phenol concentration (0.125, 0.25 and 0.5 mM). An accumulationof five scans with a scan speed of 50 nm per minute was performedand data were collected for each nm from 260 to 180 nm. Sampletemperature was maintained at 25 �C using a Neslab RTE-111 cir-culating water bath connected to the water-jacketed quartz cuv-ettes. Spectra were corrected for buffer signal and conversion tothe Mol CD (De) was performed with the Jasco Standard Analysissoftware. The protein secondary structure was calculated usingCDSSTR, which calculates the different assignments of secondarystructures by comparison with CD spectra, measured from differ-ent proteins for which high quality X-ray diffraction data are avail-able (Johnson, 1999; Sreerama & Woddy, 2000). The programCDSSTR is provided in CDPro software package which is availableat the website: http://lamar.colostate.edu/~sreeram/CDPro.
2.6. Fluorescence spectroscopy
Fluorometric experiments were carried out on a Perkin–ElmerLS55 Spectrometer. Stock solutions of polyphenol 1 mM were pre-pared at room temperature (24 ± 1 �C). Various solutions of poly-phenol (20–400 lM) were prepared from the above stocksolutions by successive dilutions also at 24 ± 1 �C. Solution of b-lac-toglobulin (200 lV) in 10 mM Tris–HCl (pH 7.4) was also prepared
Table 1Secondary structure analysis (infrared spectra) for the free b-lactoglobulin and its tea polyphenol complexes in hydrated film at pH 7.4.
Amide I components (cm�1) Free b-LG (%) 0.25 mM C–b-LG (%) 0.5 mM EC–b-LG (%) 0.5 mM ECG–b-LG (%) 0.5 mM EGCG–b-LG (%) 0.5 mM
1692–1680 b-anti 23 ± 1 13 ± 1 15 ± 1 6 ± 1 7 ± 11680–1660 turn 14 ± 1 20 ± 3 13 ± 1 16 ± 1 13 ± 11660–1650 a-helix 12 ± 1 14 ± 3 15 ± 1 13 ± 2 15 ± 11640–1610 b-sheet 51 ± 1 53 ± 2 57 ± 1 65 ± 2 64 ± 2
Table 2Secondary structure of b-lactoglobulin complexes (CD spectra) with tea polyphenol at pH 7.4. Calculated by CDSSTR software.
Polyphenol concentration (mM) a-Helix (±1%) b-Sheet (±3%) Turn (±1%) Random (±2%)
Free b-LG 12 50 12 26Catechin–b-LG (0.5 mM) 12 54 15 19Epicatechin–b-LG (0.5 mM) 13 52 15 20ECG–b-LG (0.5 mM) 14 56 14 16EGCG–b-LG (0.5 mM) 14 57 15 14
Fig. 3. Spectral changes for b-lactoglobulin CH2 symmetric and antisymmetric stretching vibrations upon polyphenol complexation (the contribution from free polyphenolvibrations has been subtracted in this region).
1050 C.D. Kanakis et al. / Food Chemistry 127 (2011) 1046–1055
C.D. Kanakis et al. / Food Chemistry 127 (2011) 1046–1055 1051
at 24 ± 1 �C. The above solutions were kept in the dark and usedsoon after. Samples containing 0.4 ml of the above protein solutionand various polyphenol solutions were mixed to obtain final poly-phenol concentration of 10–200 lV with constant b-LG content100 lV. The fluorescence spectra were recorded at kexc = 280 nmand kemi from 290 to 500 nm. The intensity at 340 nm (tryptophan)
Fig. 4. Fluorescence emission spectra of polyphenol-b-lactoglobulin systems in10 mM Tris–HCl buffer pH 7.4 at 25 �C for A (C–b-LG) (1) free b-LG 100 lM, (2–12) Cat 10, 20, 40, 60, 80, 100, 120, 140, 160, 180, 200 lM; B (EC–b-LG) (1) free b-LG100 lM, (2–12) EC at 10, 20, 40, 60, 80, 100, 120, 140, 160, 180, 200 lM; C (ECG–b-LG) (1) free b-LG 100 lM, (2–12) ECG at 10, 20, 40, 60, 80, 100, 120, 140, 160, 180,200 lM; D (EGCG–b-LG) (1) free b-LG 100 lM, (2–12) EGCG at 10, 20, 40, 60, 80,100, 120, 140, 160, 180, 200 lM. The plot of F0/(F0 � F) as a function of 1/polyphenolconcentration. The binding constant K being the ratio of the intercept and the slopefor polyphenol–b-LG complexes (A0–H0).
was used to calculate the binding constant (K) according to previ-ous literature reports (Bi et al., 2004; Dufour & Dangles, 2005; Heet al., 2005; Tang, Qi, & Chen, 2005).
2.7. Docking studies
The docking studies were performed with ArgusLab 4.0.1 soft-ware (Mark A. Thompson, Planaria Software LLC, Seattle, WA,http://www.arguslab.com). The b-LG structures were obtainedfrom the literature (Qin et al., 1998) and the polyphenol threedimensional structures were generated from PM3 semi-empiricalcalculations using Chem3D Ultra 6.0. The whole protein was se-lected as a potential binding site since no prior knowledge of suchsite was available. The docking runs were performed on the Argus-Dock docking engine using regular precision with a maximum of1000 candidate poses. The conformations were ranked using theAscore scoring function, which estimates the free binding energy.Upon location of the potential binding sites, the docked complexconformations were optimised using a steepest decent algorithmuntil convergence, with a maximum of 20 iterations. Amino acidresidues within a distance of 3.5 Å relative to the polyphenol wereinvolved in the complexation.
3. Results and discussion
3.1. FTIR spectra of polyphenol–b-LG complexes
The polyphenol–b-lactoglobulin complexation was character-ised by infrared spectroscopy and its derivative methods. Sincethere was no major spectral shifting for the protein amide I bandat 1660 cm�1 (mainly C@O stretch) and amide II band at1530 cm�1 (C–N stretching coupled with N–H bending modes)(Beauchemin et al., 2007; Krimm & Bandekar, 1986) upon polyphe-nol interaction, the difference spectra [(protein solution + polyphe-nol solution) � (protein solution)] were obtained, in order tomonitor the intensity variations of these vibrations and the resultsare shown in Fig. 1. Similarly, the infrared self-deconvolution withsecond derivative resolution enhancement and curve-fitting proce-dures (Byler & Susi, 1986) were used to determine the protein sec-ondary structures in the presence of polyphenols (Fig. 2 andTable 1).
At low polyphenol concentration (0.125 mM), an increase inintensity was observed for the protein amide I at 1660 and amideII at 1530 cm�1, in the difference spectra of the polyphenol–b-LGcomplexes (Fig. 1, diffs., 0.125 mM). The positive features arelocated in the difference spectra for amide I and II bands at 1661,1516 (C–b-LG), 1650, 1514 (EC–b-LG), 1631, 1536 (ECG–b-LG)and at 1630, 1538 cm�1 (EGCG–b-LG) in the polyphenol–b-LGcomplexes. Similarly a positive broad feature at 1628–1608 cm�1
in the difference spectra of polyphenol–protein complexes comesfrom a major increase in intensity of the band at 1628 cm�1 relatedto the protein b-sheet component (Fig. 1, diff., 0.125 mM). Thesepositive features are related to an increase of the intensity of theamide I and amide II bands upon polyphenol–protein complexa-tion. The increase in intensity of the amide I and amide II bands
Table 3Calculated binding constants (K) for the polyphenol–b-lactoglobulin complexes andthe number of bound polyphenol (n) per b-lactoglobulin molecule.
Complexes K-Fluorescence (103 M�1) n
C–b-LG 2.2 ± 0.8 1.1 ± 0.04EC–b-LG 3.2 ± 1.0 0.9 ± 0.08ECG–b-LG 11 ± 0.6 0.9 ± 0.07EGCG–b-LG 13.4 ± 0.8 1.3 ± 0.06
Fig. 5. The plot of log(F0 � F)/F as a function of log [polyphenol] for calculation ofnumber of binding (n) in polyphenol–b-LG complexes.
1052 C.D. Kanakis et al. / Food Chemistry 127 (2011) 1046–1055
is due to polyphenol binding to protein C@O, C–N and N–H groups(hydrophilic interaction). Additional evidence to support thepolyphenol interaction with C–N and N–H groups comes fromthe shifting of the protein amide A band at 3300 cm�1 (N–Hstretching mode) in the free b-LG to 3290–3280 cm�1, upon poly-phenol interaction (spectra not shown).
As polyphenol concentration increased to 0.5 mM, an increasein the intensity of the protein amide I and amide II bands wasobserved with positive features in the difference spectra foramide I and II bands at 1665, 1513 (C–b-LG), 1654, 1515 (EC–b-LG), 1630, 1540 (ECG–b-LG) and at 1630, 1532 cm�1 (EGCG–b-LG) upon polyphenol complexation (Fig. 1, diff., 0.5 mM). Thefurther increase in intensity of the amide I band in the spectraof the polyphenol–b-LG complexes suggests an increase in pro-tein a-helical and b-sheet structures at high polyphenol concen-trations. Similar infrared spectral changes observed for proteinamide I band in several ligand–protein complexes, where majorprotein conformational changes occurred (Ahmed Ouameuret al., 2006).
A quantitative analysis of the protein secondary structure forthe free b-LG and its polyphenol adducts in hydrated films hasbeen carried out and the results are shown in Fig. 2 and Table 1.The free b-LG has major b-sheet content 51% (1633, 1618), a-helix12% (1653 cm�1), turn 19% (1668 cm�1) and b-antiparallel 23%(1687 cm�1) (Fig. 2 and Table 1). These data are consistent withspectroscopic studies of b-LG previously reported (Ql et al.,1997). Upon polyphenol interaction, a major increase of b-sheetand a minor increase of a-helix was observed, upon polyphenol–b-LG complexation (Table 1). The conformational changes ob-served were more pronounced for ECG and EGCG than C and ECcomplexes (Fig. 2 and Table 1). This is indicative of larger perturba-tions of protein secondary structure by larger and bulkier polyphe-nols. This is also consistent with the extra stability of b-LG uponpolyphenol interaction.
3.2. CD spectroscopy
CD spectroscopy was also used to analyse the protein confor-mation in the polyphenol–b-LG complexes and the results areshown in Table 2. The CD results exhibit marked similarities withthose of the infrared data (Table 2). The protein conformationalanalysis based on CD data suggests that free b-LG has a-helical12%, b-sheet 50%, turn 12% and random coil 26% (Table 2), consis-tent with the literature report (Liang, Tajmir-Riahi, & Subirade,2008; Ql et al., 1997). Upon polyphenol interaction, an increaseof b-sheet from 50% (free protein) to 54–57% and a-helix from12% (free protein) to 13–14% in the polyphenol–b-LG complexes(Table 2). The increase of a-helix and b-sheet structure wasaccompanied by the reduction of random coil from 26% to 19–14%, upon polyphenol complexation (Table 2). The increase in b-sheet and a-helix contents and decrease in random coil are indic-ative of protein structural stabilisation in the presence of tea poly-phenols, which is consistent with our infrared analysis (Tables 1and 2).
3.3. Hydrophobic interactions
The spectral changes of the protein CH2 antisymmetric andsymmetric stretching vibrations, in the region of 3000–2800 cm�1 were monitored in order to locate the presence ofhydrophobic contact in the polyphenol–b-lactoglobulin complexes.The CH2 bands of the free b-lactoglobulin at 2945, 2917, 2895 and2833 cm�1 shifted to 2946, 2912 and 2897 cm�1 (C–b-LG), to 2943,2915, 2893 and 2835 cm�1 (EC–b-LG), to 2946, 2919, 2897 and2835 cm�1 (ECG–b-LG) and to 2947, 2919, 2897 and 2835 cm�1
(EGCG–b-LG), upon polyphenol complexation (Fig. 3). The shifting
of the protein antisymmetric and symmetric CH2 stretching vibra-tions suggests the presence of hydrophobic interactions via poly-phenol rings and hydrophobic pockets in b-LG, which isconsistent with fluorescence spectroscopic results discussedbelow.
C.D. Kanakis et al. / Food Chemistry 127 (2011) 1046–1055 1053
3.4. Fluorescence spectra and stability of polyphenol–casein complexes
b-Lactoglobulin has two tryptophan residues Trp-19 and Trp-61. Trp-19 is in an apolar environment and contributes to 80% oftotal fluorescence, whilst Trp-61 is partly exposed to aqueous sol-vent and has a minor contribution to Trp fluorescence (Liang et al.,2008). When other molecules interact with b-LG, tryptophan fluo-rescence may change depending on the impact of such interactionon the protein conformation (Lakowicz, 1999; Tayeh et al., 2009).On the assumption that there are (n) substantive binding sitesfor quencher (Q) on protein (B), the quenching reaction can beshown as follows:
nQ þ B() Q nB ð1Þ
The binding constant (KA), can be calculated as:
KA ¼ ½Q nB�=½Q �n½B� ð2Þ
where, [Q] and [B] are the quencher and protein concentration,respectively, [QnB] is the concentration of non fluorescent fluoro-
Fig. 6. Best docked conformations of tea polyphenols–b-LG complexes. Residues of interecomplexed to b-LG, (B) for epicatechin complexed to b-LG and (C) for epigallocatechin galegend, the reader is referred to the web version of this article.)
phore–quencher complex and [B0] gives total proteinconcentration:
½QnB� ¼ ½B0� � ½B� ð3ÞKA ¼ ð½B0� � ½B�Þ=½Q �n½B� ð4Þ
The fluorescence intensity is proportional to the protein concentra-tion as described:
½B�=½B0� / F=F0 ð5Þ
Results from fluorescence measurements can be used to esti-mate the binding constant of polyphenol–protein complex. FromEq. (4):
log½ðF0 � FÞ=F� ¼ logKA þ nlog½Q � ð6Þ
The accessible fluorophore fraction (f) can be calculated bymodified Stern–Volmer equation:
F0=ðF0 � FÞ ¼ 1=fK½Q � þ 1=f ð7Þ
st are shown in red colour and the tea polyphenols in green colour; (A) for catechinllate complexed to b-LG. (For interpretation of the references to colour in this figure
Table 4Amino acid residues involved in cathechin–b-lactoglobulin interaction with the free binding energy for the best selected docking positions.
Complex Residues involved in the interaction DGbinding
(k cal/mol)
Catechin–b-LG Ala-86a, Asn-88a, Asn-90a, Asp-85, Glu-62a, Ile-71, Ile-84, Leu-39, Leu-58, Lys-60a, Lys-69, Met-107, Val-41 �8.04Epicatechin–b-LG Asn-90, Asp-85a, Ile-56, Ile-71, Ile-84, Leu-39, Leu-58, Lys-60a, Lys-69, Met-107, Phe-105, Pro-38a, Val-41 �8.28EGCG–b-LG Asn-88a, Asn-90a, Asp-85a, Gln-120, Ile-56, Ile-71, Ile-84, Leu-39, Leu-58, Lys-60, Lys-69, Lys-70a, Met-107, Phe-105, Pro-38a, Val-41 �8.16
a Hydrogen bonding with this residue.
1054 C.D. Kanakis et al. / Food Chemistry 127 (2011) 1046–1055
where, F0 is the initial fluorescence intensity and F is the fluores-cence intensities in the presence of quenching agent (or interactingmolecule). K is the Stern–Volmer quenching constant, [Q] is the mo-lar concentration of quencher and f is the fraction of accessible fluo-rophore to a polar quencher, which indicates the fractionalfluorescence contribution of the total emission for an interactionwith a hydrophobic quencher (Lakowicz, 1999). The plot of F0/(F0 � F) versus 1/[Q] yields f�1 as the intercept on y axis and(fK)�1 as the slope. Thus, the ratio of the ordinate and the slopegives K. The decrease of fluorescence intensity of b-LG is monitoredat 340 nm for polyphenol–b-LG systems (Fig. 4A–D shows represen-tative results for each system). The plot of F0/(F0 � F) versus 1/[pol-yphenol] (Fig. 4A0–D0 shows representative plots) is shown in Fig. 4.Assuming that the observed changes in fluorescence come from theinteraction between polyphenols and protein, the quenchingconstant can be taken as the binding constant of the complexformation. The K values given here are averages of three-replicateruns for polyphenol–b-LG systems, each run involving severaldifferent concentrations of polyphenol (Fig. 4). KC–b-LG = 2.2(±0.8) � 103 M�1, KEC–b-LG = 3.2 (±1) � 103 M�1, KECG–b-LG = 1.1(±0.6) � 104 M�1 and KEGCG–b-LG = 1.3 (±0.8) � 104 M�1 (Fig. 4A0–D0
and Table 3). The binding constants calculated for the polyphe-nol–b-LG suggest a low affinity polyphenol–b-LG interaction, com-pared to the other strong ligand–protein complexes (Kragh-Hansen,1990; Kratochwil, Huber, Muller, Kansy, & Gerber, 2002; N’soukpoé-Kossi et al., 2007). However, similar binding constants (103–104 M�1) were also reported for several ligand–protein complexesusing fluorescence spectroscopic methods (Bi et al., 2004; Lianget al., 2008; Sulkowska, 2002). The binding constants of the largerpolyphenol–b-LG complexes are bigger than those of smaller poly-phenol–b-LG adducts, which can be due to the presence of more OHgroups associated with the bulkier polyphenols (Table 3).
The number of polyphenols bound per protein (n) is calculatedfrom log [(F0 � F)/F] = logKS + n log [polyphenol] for the staticquenching (Charbonneau & Tajmir-Riahi, 2010; Froehlich,Jennings, Sedaghat-Herati, & Tajmir-Riahi, 2009; Jiang, Gao, & He,2002; Jiang et al., 2004; Liang et al., 2008; Mandeville &Tajmir-Riahi, 2010). The n values from the slope of the straight lineplot in Fig. 5 are for polyphenol–b-LG complexes 1.1 (C), 0.9 (EC),0.90 (ECG), 1.3 (EGCG) (Fig. 5 and Table 3).
The f values obtained for polyphenol–b-LG complexes suggestthat polyphenols interact with fluorophore via both hydrophobicand hydrophilic interactions. As a result, we predict that polyphe-nols bind mainly with the fluorophores located inside of b-LG. Thisargument is based on the fact that the emissions kmax of Trp-214(HSA) and Trp-212 (BSA) are at 340 nm, which is the emission re-gion of hidden tryptophan molecules, whilst fluorescence emissionof exposed tryptophan molecule is at a higher wavelength(350 nm) due to solvent relaxation (Bourassa, Kanakis, Tarantilis,Polissiou, & Tajmir-Riahi, 2010; Liang et al., 2008; Sulkowska,2002; Tayeh et al., 2009). The tightening of protein structurethrough intramolecular interactions, such as hydrogen bondsseems to bury tryptophan in a more hydrophobic environment.The changes in fluorescence intensity of tryptophan in b-LG in
the presence of polyphenol exhibited two different patterns. Inthe catechin, epicatechin and epicatechin gallate complexes of b-LG, the emission band of the free protein at 340 nm (b-LG) shiftedtowards a lower wavelength at 335 nm (Fig. 4A–C), whilst for theEGCG complex it was observed at a higher wavelength of 345 nm(Fig. 4D). The downward shift of the emission band of b-LG inthe catechin, epicatechin and epicatechin gallate–protein com-plexes is due to the tightening of protein structure through intra-molecular interactions, such as hydrogen bonds, appears tointroduce tryptophan in a more hydrophobic environment. How-ever the upward shift of the protein emission bands observed inthe spectra of EGCG–b-LG is related to more exposure of trypto-phan residue and unfolding of protein structure.
3.5. Docking study
Our results from FTIR, CD and fluorescence spectroscopic meth-ods are accompanied by docking experiments in which the C, ECand EGCG molecules were docked to b-lactoglobulin to determinethe preferred binding sites on this protein. The stereoview of thedockings of C, EC and EGCG are shown in Fig. 6 and Table 4. Thedocking results show that C is surrounded by Ala-86, Asn-88,Asn-90, Asp-85, Glu-62, Ile-71, Ile-84, Leu-39, Leu-58, Lys-60,Lys-69, Met-107 and Val-41, whilst EC is in the vicinity of Asn-90, Asp-85, Ile-56, Ile-71, Ile-84, Leu-39, Leu-58, Lys-60, Lys-69,Met-107, Phe-105, Pro-38 and Val-41 (Fig. 6 and Table 4). EGCGis in the vicinity of Asn-88, Asn-90, Asp-85, Gln-120, Ile-56, Ile-71, Ile-84, Leu-39, Leu-58, Lys-60, Lys-69, Lys-70, Met-107, Phe-105, Pro-38 and Val-41 (Fig. 6 and Table 4). The binding energy(DG) shows more stable polyphenol–protein complexes formedwith epicatechin and epigallocatechin gallate than catechin (Ta-ble 4). Hydrogen bindings were observed between C, EC and EGCGand different amino acid residues that stabilise polyphenol–pro-tein complexes (Table 4).
4. Conclusions
The spectroscopic results and docking studies presented hereshow, tea polyphenols weakly bind to b-lactoglobulin in solution.The order of binding increases as the number of OH groupincreased with EGCG > ECG > EC > C. Both hydrophobic andhydrophilic interactions are observed in the polyphenol–b-lacto-globulin complexation. Polyphenol binding alters protein second-ary structure with an increase in b-sheet and a-helix, leading toprotein structural stabilisation. The b-LG structural changes canbe a major factor in the effect of milk on the antioxidant capacityof tea polyphenols.
Acknowledgments
The financial support of the Natural Sciences and EngineeringResearch Council of Canada (NSERC) to H.A. Tajmir-Riahi is highlyappreciated and also the scientific support from the AgriculturalUniversity of Athens, Greece.
C.D. Kanakis et al. / Food Chemistry 127 (2011) 1046–1055 1055
References
Aguie-Beghin, V., Sausse, P., Meudec, E., Cheynier, V., & Doullard, R. (2008).Polyphenol–b-casein complexes at air/water interface and in solution: Effects ofpolyphenol structure. J. Agric. Food Chem., 56, 9600–9611.
Ahmed Ouameur, A., Diamantoglou, S., Sedaghat-Herati, M. R., Nafisi, Sh.,Carpentier, R., & Tajmir-Riahi, H. A. (2006). An overview of drug binding tohuman serum albumin: Protein folding and unfolding. Cell Biochem. Biophys., 45,203–213.
Alexandropoulou, I., Komaitis, M., & Kapsokefalou, M. (2006). Effects of iron,ascorbate, meat and casein on the antioxidant capacity of green tea underconditions of in vitro digestion. Food Chem., 94, 359–365.
Beauchemin, R., N’ soukpoe-Kossi, C. N., Thomas, T. J., Thomas, T., Carpentier, R., &Tajmir-Riahi, H. A. (2007). Polyamine analogues bind human serum albumin.Biomacromolecules, 8, 3177–3183.
Bi, S., Ding, L., Tian, Y., Song, D., Zhou, X., Liu, X., et al. (2004). Investigation of theinteraction between flavonoids and human serum albumin. J. Mol. Struct., 703,37–45.
Bourassa, P., Kanakis, D. C., Tarantilis, P., Polissiou, M. G., & Tajmir-Riahi, H. A.(2010). Resveratrol. Genistein and curcumin bind bovine serum albumin. J. Phys.Chem. B, 114, 3348–3354.
Byler, D. M., & Susi, H. (1986). Examination of the secondary structure of proteins bydeconvoluted FTIR spectra. Biopolymers, 25, 469–487.
Charbonneau, D. M., & Tajmir-Riahi, H. A. (2010). Study on the interaction ofcationic lipids with bovine serum albumin. J. Phys. Chem. B, 114, 1148–1155.
Collini, M., Alfonso, D. L., & Baldini, G. (2000). New insights on b-lactoglobulinbinding sites by 1-anilinonaphthalene-8-sulfonate fluorescence decay. ProteinSci., 9, 1968–1974.
De Fretias, N., & Mateus, S. (2001). Structural features of procyanidin interactionswith salivary proteins. J. Agric. Food Chem., 49, 940–945.
Dousseau, F., Therrien, M., & Pezolet, M. (1989). On the spectral subtraction of waterfrom the FT-IR spectra of aqueous solutions of proteins. Appl. Spectrosc., 43,538–542.
Dubeau, S., Samson, G., & Tajmir-Riahi, H. A. (2010). Duel effect of milk on theantioxidant capacity of green, Darjeeling, and English breakfast teas. FoodChem., 122, 539–545.
Dufour, C., & Dangles, O. (2005). Flavanoid–serum albumin complexation:Determination of binding constants and binding sites by fluorescencespectroscopy. Biochim. Biophys. Acta, 1721, 164–173.
Froehlich, E., Jennings, C. J., Sedaghat-Herati, M. R., & Tajmir-Riahi, Y. F. (2009).Dendrimers bind human serum albumin. J. Phys. Chem. B, 113, 6986–6993.
Hasni, I., Bourassa, P., Hamdani, S., Samson, G., Carpentier, R., & Tajmir-Riahi, H. A.(2011). Interaction of milk a- and b-caseins with tea polyphenol. Food. Chem.,126, 630–639.
He, W., Li, Y., Xue, C., Hu, Z., Chen, X., & Sheng, F. (2005). Effect of Chinese medicinealpinetin on the structure of human serum albumin. Bioorg. Med. Chem., 13,1837–1845.
Jiang, C. Q., Gao, M. X., & He, J. X. (2002). Study of the interaction between terazosinand serum albumin synchronous fluorescence determination of terazosin. Anal.Chim. Acta, 452, 185–189.
Jiang, M., Xie, M. X., Zheng, D., Liu, Y., Li, X. Y., & Chen, X. (2004). Spectroscopicstudies on the interaction of cinnamic acid and its hydroxyl derivatives withhuman serum albumin. J. Mol. Struct., 692, 71–80.
Jobstl, E., Howse, J. R., Fairclough, J. P. A., & Williamson, M. P. (2006). Nocovalentcross-linking of casein by epigallocatechingallate characterized by singlemolecule force microscopy. J. Agric. Food Chem., 54, 4077–4081.
Johnson, W. C. (1999). Analyzing protein circular dichroism spectra for accuratesecondary structure. Proteins Struct. Funct. Genet., 35, 307–312.
Kartsova, L. A., & Alekseeva, A. V. (2008). Effect of milk caseins on the concentrationof polyphenolic compounds in tea. J. Anal. Chem., 63, 1107–1111.
Kragh-Hansen, U. (1990). Structure and ligand binding properties of human serumalbumin. Dan. Med. Bull., 37, 57–84.
Kratochwil, N. A., Huber, W., Muller, F., Kansy, M., & Gerber, P. R. (2002). Predictingplasma protein binding of drugs: A new approach. Biochem. Pharmacol., 64,1355–1374.
Krimm, S., & Bandekar, J. (1986). Vibrational spectroscopy and conformation ofpeptides, polypeptides, and proteins. Adv. Protein Chem., 38, 181–364.
Lakowicz, J. R. (1999). Principles of Fluorescence Spectroscopy (2nd ed.). New York:Kluwer/Plenum.
Liang, Y., & Xu, Y. (2003). Effect of extraction temperature on cream andextractibility of black tea [Camellia sinensis (L.) O. Kuntze]. Int. J. Food Sci.Technol., 38, 37–45.
Liang, L., Tajmir-Riahi, H. A., & Subirade, M. (2008). Interaction of b-lactoglobulinwith resveratrol and its biological implications. Biomacromolecules, 9, 50–55.
Liang, L., & Subirade, M. (2010). b-Lactoglobulin/folic acid complexes: Formation,characterization, and biological implication. J. Phys. Chem. B, 114, 6707–6712.
McKenzie, H. A., & Sawyer, W. H. (1967). Effect of pH on b-lactoglobulins. Nature,214, 1101–1104.
Mandeville, J. S., & Tajmir-Riahi, H. A. (2010). Complexes of dendrimers with bovineserum albumin. Biomacromolecules, 11, 465–472.
N’soukpoé-Kossi, C. N., Sedaghat-Herati, M. R., Ragi, C., Hotchandani, S., & Tajmir-Riahi, H. A. (2007). Retinol and retinoic acid bind human serum albumin:Stability and structural features. Int. J. Biol. Macromol., 40, 484–490.
Qin, B. Y., Bewley, M. C., Creamer, L. K., Baker, H. M., Baker, E. N., & Jameson, G. B.(1998). Structural basis of the tanford transition of bovine b-lactoglobulin.Biochemistry, 37, 14014–14023.
Ql, X. L., Holt, C., Mcnulty, D., Clarke, D. T., Brownlow, S., & Jones, G. R. (1997). Effectof temperature on the secondary structure of b-lactoglolobulin at pH 6.7, asdetermined by CD and IR spectroscopy: A test of the molten globule hypothesis.Biochem. J., 324, 341–346.
Shukla, A., Narayanan, T., & Zanchi, D. (2009). Structure of casein micelles and theircomplexation with tannins. Soft. Matter, 5, 2884–2888.
Sreerama, N., & Woddy, R. W. (2000). Estimation of protein secondary structurefrom circular dichroism spectra: Comparison of CONTIN, SELCON and CDSSTRmethods with an expanded reference set. Anal. Biochem., 287, 252–260.
Stanner, S. (2007). Does adding milk remove the benefit of your daily cuppa? Br.Nutr. Foundation Nutr. Bull., 32, 101–103.
Sulkowska, A. (2002). Interaction of drugs with bovine and human serum albumin.J. Mol. Struct., 614, 227–232.
Tang, J., Qi, S., & Chen, X. (2005). Spectroscopic studies of the interaction of anti-coagulant rodenticide diphacinone with human serum albumin. J. Mol. Struct.,779, 87–95.
Tayeh, N., Rungassamy, T., & Albani, J. R. (2009). Fluorescence spectral resolution oftryptophan residues in bovine and human serum albumins. J. Pharm. Biomed.Anal., 50, 107–116.
Yan, Y., Hu, J., & Yao, P. (2009). Effects of casein, ovalbumin, and dextran on theastringency of tea polyphenols determined by quartz crystal microbalance withdissipation. Langmuir, 25, 397–402.