Food Research International 53 (2013) 449455

Contents lists available at SciVerse ScienceDirect

Food Research International

journal homepage: www.elsevier.com/locate/foodres

Interactions of black and green tea polyphenols with whole milk

Jianhui Ye a,b, Fangyuan Fan a, Xinqing Xu b, Yuerong Liang a,
a
b

Zhejiang University Tea Research Institute, Hangzhou 310058, China

CSIRO Animal, Food and Health Sciences, 671 Sneydes Road, Werribee, VIC 3030, Australia

a r t i c l e

i n f o

Article history:
Received 19 February 2013
Accepted 26 May 2013
Keywords:
Tea polyphenols
Caseins
FTIR
Fluorescence quenching
ORAC assay

a b s t r a c t
Interactions of black tea polyphenols (BTP) and green tea polyphenols (GTP) with whole milk were comparatively studied. Upon the ultracentrifugation of BTP and GTPmilk systems (40% v/v milk), most BTP and
GTP partitioned into milk protein fractions: whey proteins and casein micelles, with 35.1% and 73.0% of
total catechins bound to the casein micelles accordingly. The afnities of catechins for casein micelles were
differentiated by the structures of catechins in the GTPmilk system but the BTPmilk system, being enhanced by a gallate group in catechins and the cis-form and weakened by a pyrogallol group. Fourier transforms infrared spectroscopy (FTIR) analysis showed that TP binding altered the secondary structures of
milk proteins by reducing inter -sheet, random coil and the large loop and increasing -helix, intra
-sheet and turn structures, and more intense hydrophobic interaction was observed in the BTPmilk system. UVvis spectra indicated no obvious impacts on TP molecules at a low concentration of milk proteins.
With the increasing addition of tea infusion, BTP exhibited a similar uorescence quenching ability to GTP,
but the variance in the ORAC values of BTPmilk system was different from that of the GTPmilk system,
suggesting that uorescence quenching may not fully represent the interactions between polyphenols and
proteins.
2013 Elsevier Ltd. All rights reserved.

1. Introduction
Tea is a worldwide beverage having various health benets and
physiological functionalities, such as antioxidative (Benzie & Szeto,
1999), anticarcinogenic and antimutagenic effects (Dufresne &
Farnworth, 2000; Gupta, Saha, & Giri, 2002), due to the principal
bioactive components tea polyphenols (TP). Tea catechins are
the major components of TP, which consist of ()-epigallocatechin
gallate (EGCg), ()-epigallocatechin (EGC), ()-epicatechin
gallate (ECg), ()-epicatechin (EC), and their epimerization
isomers (+)-gallocatechin gallate (GCg), (+)-gallocatechin (GC),
(+)-catechin gallate (Cg), and (+)-catechin (C) (Goto, Yoshida, Kiso,
& Nagashima, 1996). Black tea (fully fermented) and green tea
(non-fermented) are two major tea varieties with different chemical
components named black tea polyphenols (BTP) and green tea polyphenols (GTP). Consumption of black tea with milk is a regular practice
in daily life, and the application of tea or tea extract to dairy product is
becoming popular thanks to the antibacterial effect and bioactivities
of TP (Ferruzzi & Green, 2006). Studies have been extensively carried
out concerning the impact of the addition of milk or milk proteins on
the antioxidant potentials of TP (Arts et al., 2002; Dubeau, Samson, &
Tajmir-Riahi, 2010; Langley-Evans, 2000b; Lorenz et al., 2007; Serani,
Ghiselli, & FerroLuzzi, 1996), and three types of results were achieved:

Corresponding author. Tel./fax: +86 571 88982704.

E-mail address: yrliang@zju.edu.cn (Y. Liang).
0963-9969/$ see front matter 2013 Elsevier Ltd. All rights reserved.
http://dx.doi.org/10.1016/j.foodres.2013.05.033

non-masking effect of adding milk or milk proteins (Kyle, Morrice,

McNeill, & Duthie, 2007; Leenen, Roodenburg, Tijburg, & Wiseman,
2000; Richelle, Tavazzi, & Offord, 2001), differentiated masking effect
on the antioxidant potentials of black tea and green tea (Arts et al.,
2002; Dubeau et al., 2010), and total inhibition on both (Serani et al.,
1996). Understanding the interactions of BTP and GTP with milk is important to clarify the controversies over milk impacts on the antioxidant
activities of black tea and green tea.
Whole milk contains 85.588.7% water, 2.34.4% protein and 2.4
5.5% fat. Caseins and whey proteins are the major proteins in milk,
which account for 80% and 20% of milk proteins. Caseins exist in the
form of casein micelles with hydrophilic surface layer and hydrophobic interior (Sahu, Kasoju, & Bora, 2008). -Lactoglobulin is the major
component of whey proteins. Due to the rich contents in proteins and
lipids, milk or milk isolates have been used as carriers for bioactive
compounds via ligand binding in order to improve the sensation,
water solubility or bioavailability of bioactives (Bohin, Vincken, van
der Hijden, & Gruppen, 2012; Livney, 2010; Staszewski et al., 2012).
The interactions between individual catechins (e.g. EGCg, EGC and
ECg) and pure proteins (e.g. -casein, -casein and -lactoglobulin)
or milk proteins occurred with the formation of catechinprotein
complexes, and EGCg had stronger afnities for the pure proteins or
milk proteins compared with C, EC and EGC (Bohin et al., 2012;
Hasni et al., 2011; Kanakis et al., 2011; Xiao et al., 2011). However,
the behaviors of tea catechins might be different in the compound
milk system reconstituted by BTP or GTP rather than individual catechins, and the relevant information is scarce.

450

J. Ye et al. / Food Research International 53 (2013) 449455

Fluorescence spectroscopy has been widely used to interpret the interactions between polyphenols and proteins (Bourassa, Kanakis,
Tarantilis, Pollissiou, & Tajmir-Riahi, 2010; Hemar, Gerbeaud, Oliver, &
Augustin, 2011), and the oxygen radical absorbance capacity (ORAC)
assay is a common method for measuring the antioxidant capacities of
polyphenols and milk (Huang, Ou, Hampsch-Woodill, Flanagan, &
Prior, 2002; Zulueta, Esteve, & Frigola, 2009; Zulueta, Maurizi, Frigola,
Esteve, Coli and Burini, 2009). Normally, the interaction between polyphenol and protein is assumed as the main reason for protein impacts
on the antioxidant activities of polyphenols, but the correlation between them is still equivocal. In this study, the afnities of tea catechins
for different components of milk in both the BTP and GTPmilk systems were investigated by using ultracentrifugation separation. Fourier
transforms infrared spectroscopy (FTIR) and UVvis spectroscopic
methods were used to investigate the impacts of TPprotein interactions on the secondary structures of milk proteins and the molecules
of TP. Protein uorescence quenching study was carried out to interpret
the interactions between TP and milk proteins, and the correlations between the uorescence quenching ratio% and the variances in the ORAC
values of TPmilk systems were analyzed.
2. Materials and methods
2.1. Materials and chemicals
Green tea and black tea were supplied by the CinoTea CO., Ltd.
(Hangzhou, China). Ultra high temperature processed (UHT) full
cream milk (3.2 g protein, 3.3 g fat, 120 mg calcium per 100 mL)
was purchased from the local supermarket. Phosphate buffer (PB),
6-hydroxy-2,5,7,8-tetramethylchroman-2-carboxylic acid (Trolox),
uorescein, 2,2-azobis (2-methylpropionamidine) dihydrochloride
(AAPH), sodium carbonate, catechins (C), FolinCiocalteu reagent,
HPLC standards (EGCg, EGC, ECg, EC, GCg, GC, Cg and C) as well as
other chemical reagents of HPLC grade were purchased from SigmaAldrich (Sydney, Australia).
2.2. Preparation of tea infusion stocks
Green tea and black tea were ground by Breville CG2B grinder
(Breville Group Ltd., Australia) and sifted through 0.45 mm sifter respectively. The ground tea samples (10.0 g) were brewed with
300 mL boiling water for 3 min. After ltering through the silicone
treated lter paper (Whatman International Ltd., Maidstone, England),
the tea infusions were centrifuged at 10,000 g and 20 for 45 min.
The obtained supernatants were kept at 4 , namely black tea stock
(pH 5.18) and green tea stock (pH 5.82).
2.3. Partition of BTP and GTPmilk systems
Milk tea is a common object in the study of milk impacts on antioxidant potentials of tea in vivo and in vitro, thus a milk tea formula with
the addition of 40% v/v milk was applied to investigate the distributions
of tea catechins in the BTP and GTPmilk systems according to Sharma,
Kumar, and Rao (2008). 3 mL tea infusion stock was mixed with 2 mL
whole milk (pH 6.63) or 2 mL water (control), and the mixture was
kept at room temperature 20 C for 1 h. The obtained BTP and GTP
milk systems (pH 6.50 and 6.60) as well as the tea infusion controls
were ultracentrifuged (Beckman Coulter Optima L-90K ultracentrifuge)
at 367,290 g and 20 C for 1 h, with no obvious effect on the
components of ultracentrifuged fractions over the pH range of 6.56.7
at 20 C (Anema & Klostermeyer, 1997). Three fractions were obtained
from the TPmilk system: the cream layer (mainly milk fat globule, the
top layer), the skimmed serum (mainly whey proteins, the middle
layer) and the casein micelle pellet (mainly casein micelles, the bottom
layer) (Garcia-Risco, Ramos, & Lopez-Fandino, 2002). The cream layer
and the skimmed serum were collected respectively, and the skimmed

serum was made up to 5 mL using water, same volume as before

ultracentrifugation.
TP was extracted from the cream layer and the adjusted skimmed
serum with ethanol. 200 L sample was mixed with 800 L ethanol to
achieve a nal concentration of 80% v/v ethanol so as to denature proteins. After centrifugation at 10,000 g and 20 C for 15 min, the protein precipitate was re-extracted twice with fresh 80% ethanol, and
the supernatants from the same sample were combined and made
up to 5 mL for HPLC analysis and total phenolics analysis.
HPLC analysis of individual catechins: injection volume 10 L,
VYDAC C18 monomeric 238EV52, column temperature 28 , mobile
phase A = acetonitrile + 0.08% triuoroacetic acid, mobile phase
B = water + 0.1% triuoroacetic acid, linear gradient elution: from
2% (v) A/98% (v) B to 24% (v) A/76% (v) B during early 45 min and
then 2% (v) A/98% (v) B till 50 min, ow rate 200 L min1. The eluate
was monitored by Surveyor PDA Detector at 280 nm. Catechins were
identied and quantied by comparing the retention time and the
peak area with the external standards with known concentrations.
Total phenolics analysis by FolinCiocalteu assay (Komes, Horzic,
Belscak, Ganic, & Vulic, 2010): 100 L sample and 80 L 6-fold diluted
FolinCiocalteu's phenol reagent were loaded onto a 96-well microplate
and stood for 3 min. Then sodium carbonate solution (7.5% w/v, 120 L)
was added, mixed well and kept in dark for 2 h. The absorbance of the
mixture was measured by microplate reader at 765 nm. The concentration of TP was expressed as mg (+)-catechins (C) L1, using the linear
calibration curve of C (0.0050.05 mg mL1).
2.4. FTIR spectroscopic measurements
In order to diminish the interference of milk fat, the whole milk
was skimmed by centrifugation at 3000 g for 20 min (Corredig &
Dalgleish, 1996). BTP and GTPmilk samples with the addition of
40% v/v milk were reconstituted as above, using the skimmed milk.
The skimmed milk, the BTP and GTPmilk mixture samples and
the tea infusions were freeze dried (Christ Alpha 12 LD plus Freeze
Drye, German) for FTIR analysis. KBr pellet was prepared by admixing
23 mg of obtained sample with 300 mg of spectroscopy-grade KBr
and pressing the mixture into a 13-mm disk at 4 t pressure with a
die press. The spectra of pellets were recorded by a Nicolet AVATAR
370 FTIR Spectrometer (Thermo Fisher, Massachusetts, USA) from
400 to 4000 cm1.
The secondary structures of proteins were analyzed according to the
reported method (Bourassa et al., 2010). Fourier self-deconvolution and
secondary derivative were applied to the range of 17001660 cm1
assigned to the amide I band in protein FTIR spectrum. The FTIR spectra
were smoothed, and their baselines were corrected automatically using
Thermo Scientic OMNIC software. By means of the second derivative
in the spectral region of 17001600 cm1, the major peaks for protein
secondary structure were resolved. The above spectral region was
deconvoluted by the curve-tting method with the LevenbergMarquadt
algorithm and the corresponding peaks were adjusted and the area
measured with the Gaussian function: -helix (1652 2 cm1),
intermolecular -sheet (16241610 cm1), intramolecular -sheet
(16451625 cm1), turn (16851660 cm1), random coil (1648
1641 cm1), and -antiparallel (16901700 cm1). The bands located
at 1658 2 cm1 were assigned to the large loop rather than -helix
or -turns (Curley, Kumosinski, Unruh, & Farrell, 1998; Farrell,
Wickham, Unruh, Qi, & Hoagland, 2001). The area of all the component
bands assigned to a given conformation was summed up and divided
by the total area. Peak analysis was performed by the Origin Pro 8.5.1
software.
2.5. UVvis spectroscopy analysis
UVvis spectroscopy was carried out on a Shimadzu UV-1700
spectrophotometer to investigate the UVvis spectral proles of TP

J. Ye et al. / Food Research International 53 (2013) 449455

upon interaction with milk in the range of 220450 nm, using quartz
cuvettes of 1 cm. Whole milk was diluted 100-fold to decrease the
strong UVvis absorption. Tea infusion stock (100 L) was mixed
with 0, 20, 40, 80, 160, and 320 L of the 100-fold diluted whole
milk respectively, and then was adjusted to 10 mL before detection.
The UV absorption of milk was subtracted from the UVvis spectra.
2.6. Fluorescence spectroscopy
Proteins have primarily two intrinsic uorophores: tryptophan
(Trp) and tyrosine (Tyr) both excited at an ex of 280 nm (Lakowicz,
1999). Fluorometric assay was carried out on a Varioskan Flash
microplate reader (Thermo Fisher Scientic Inc., Denmark) to investigate the interaction between TP and milk proteins. Whole milk
(1200 L) was mixed with 0, 50, 100, 150, 200, 250, 300, 400, 500 and
600 L tea infusion stock respectively, and then made up to 2000 L.
300 L sample was loaded onto a 96-well microplate and the uorescence emission spectra were recorded from 300 to 450 nm with an excitation wavelength at 280 nm. The rest of the sample was submitted to
the following ORAC analysis.
2.7. ORAC assay
ORAC assay was applied to evaluate the antioxidant capacities of
the TPmilk systems according to Huang et al. (2002). The BTP
milk samples were diluted 10-fold and the GTPmilk samples were
diluted 20-fold before ORAC analysis. The diluted sample (20 L)
and 150 L uorescein (200 nM, 10 mM PB at pH 7.4) were loaded
onto a 96-well microplate and incubated for 15 min at 37 C, and
then 30 L APPH solution (240 mM, 10 mM PB at pH 7.4) was loaded
and mixed well. Fluorescence was recorded every 2 min for 200 min
at excitation and emission wavelengths of 485 and 530 nm. The area
under the curve (AUC) was calculated for each sample by integrating
the relative uorescence curve using the Origin Pro 8.5.1 software.
The net AUC of the sample was calculated by subtracting the AUC of
the blank. The nal ORAC values, expressed as M trolox L1, were
calculated by the regression equation between trolox concentration
and the net AUC using the linear calibration curve of trolox (31.25
500 M).
2.8. Data analysis
Three replicates of the BTP and GTPmilk systems were prepared
for each test and the mean values of the trials are presented. Pearson's
correlation coefcient (r) was used to evaluate the correlations between uorescence quenching ratio and masking% of ORAC values
in TPmilk systems (SPSS Statistics Software).
3. Results and discussion
3.1. Distributions of TP in the ultracentrifuged fractions of TPmilk
systems
In the study, the trace levels of tea catechins and total phenolics in
the cream layers of BTP and GTPmilk systems (data not presented)
indicated that the majority of BTP and GTP, including tea catechins,
partitioned into milk protein fractions: the skimmed serum and the
casein micelle pellet. In light of the neglectable TP in the cream
layer and various extraction efciencies of phenolics from the casein
micelle pellet, the obtained skimmed serum was made up to the
same volume as the tea infusion control, thereby the distribution ratios of tea catechins and total phenolics in the skimmed serum can
be calculated by comparing the concentrations with control, and the
approximate distribution ratios of TP in the casein micelle pellet are
obtained. Table 1 shows the concentrations and distribution ratios
of tea catechins and total phenolics in the tea infusion controls and

451

the skimmed serum fractions. For the BTP and GTPmilk systems,
the concentrations of tea catechins in the skimmed serums were
obviously lower than those in the tea infusion controls, indicating
that tea catechins in both systems were partly bound to the casein
micelles. In the BTPmilk system, the concentration of TC in the
skimmed serum was 157.9 mg L1 less than that of black tea control
(450.2 mg L1), which meant 64.9% of total catechins (TC) partitioned
into the whey protein fraction upon ultracentrifugation (Table 1). While
in the GTPmilk system, the concentration of TC in the skimmed serum
was 732.9 mg L1 less than that of green tea control (1003.5 mg L1),
indicating that only 27.0% of TC were bound to the whey proteins
(Table 1). This result suggests that the binding afnities of tea catechins
for different protein fractions are affected by the coexisting phenolics in
a compound system. Tea catechins are the important part of the phenolics in black tea and green tea. 40.4% of total phenolics were bound to
the whey proteins in the BTPmilk system, which was lower than
64.9% of TC, suggesting that more of the unknown part in BTP are
bound to the casein micelles. This is in line with the conclusion that
larger and more complex phenolics have stronger afnities for the
proline-rich proteins than smaller phenolics (Baxter, Lilley, Haslam, &
Williamson, 1997). On the contrary, 44.2% of total phenolics was preserved in the skimmed serum of the GTPmilk system compared with
27% of TC (Table 1), suggesting that more of the non-catechins part in
GTP was combined with whey proteins over casein micelles.
From Table 1, the distribution ratios of the gallated catechins
(EGCg, GCg, ECg and Cg) and the non-gallated catechins (EGC, GC,
EC and C) in the skimmed serum were obtained, being 66.5% and
63.8% for the BTPmilk system while 8.2% and 41.6% for the GTP
milk system. Likewise, the distribution ratios of the cis-type catechins
(EGC, EC, EGCg and ECg) and the trans-type catechins (GC, C, GCg and
Cg) were 63.9% and 66.4% for the BTPmilk system while 7.9% and
49.1% for the GTPmilk system. Besides, the distribution ratios of the
pyrogallol-containing catechins (GC, EGC, EGCg and GCg) and those without a pyrogallol group (C, EC, ECg and Cg) in the skimmed serum were
65.1% and 63.6% for the BTPmilk system and 32.8% and 21.1% for the
GTPmilk system. These results suggest that the interactions between casein micelles and individual catechins are not apparently related to the
molecular structures of tea catechins in the BTPmilk system but remarkably relevant in the GTPmilk system (40% v/v milk). For GTP, a gallate
group in structure and the cis-form enhance the catechincasein interaction, whereas a pyrogallol group weakens the interaction. Xiao et al.
(2011) also demonstrated that the gallated catechins had stronger binding afnities for milk proteins but the pyrogallol-containing catechins
had lower afnities in an individual catechinmilk protein system. Analogously, a higher masking extent of -casein to the gallated catechins and
a lower masking extent to the pyrogallol-containing catechins were
reported by Arts et al. (2002). EGCg is the most important catechin in
green tea leaf containing a gallate group and a pyrogallol group and in a
cis-form. The distribution ratio of EGCg in the skimmed serum of the
BTPmilk system was 65.4% much higher than that of the GTPmilk
system (7.0%, Table 1), indicating that most EGCg was bound to the casein
micelles in the GTPmilk system while the majority of EGCg was combined with whey proteins in the BTPmilk system. EGCg has a greater
afnity for -casein than -lactoglobulin under the EGCg-pure protein
circumstance (Bohin et al., 2012), but the preference of EGCg for the caseins might be moderated due to the coexistence of highly polymerized
polyphenols in the BTPmilk system.
In our study, the protein fractions incorporated most TP. The low
compatibility of TP to the lipids in the cream layer might be attributed
to the high hydrophilicity of TP and the low hydrophilicity of lipids.
Therefore TP are preferential to bind with proteins under a competitive
circumstance in the presence of proteins and lipids, although the potential of incorporating catechins to the lipids has been reported in a pure
lipid environment (Oteiza, Erlejman, Verstraeten, Keen, & Fraga,
2005). Whey proteins and casein micelles exhibited different afnities
for BTP and GTP: whey proteins preserved the majority of tea catechins

452

J. Ye et al. / Food Research International 53 (2013) 449455

Table 1
The concentrations of tea catechins and total phenolics in the tea infusion control and the skimmed serum fraction (mg L1).

BTP

Controlb
Skimmed serum

GTP

Controlb
Skimmed serum

GC

EGC

EC

EGCg

GCg

ECg

Cg

TC

135.2 1.8
90.2 1.5
(66.7%)
165.2 2.8
131.1 1.5
(79.4%)

109.2 3.5
68.3 1.7
(62.5%)
14.3 0.4
8.2 0.2
(57.3%)

7.0 0.2
3.8 0.1
(54.3%)
267.8 0.9
91.1 0.9
(34.0%)

10.8 0.2
4.9 0.2
(45.4%)
116.1 1.1
3.9 0.1
(3.4%)

120.8 2.1
79.0 0.7
(65.4%)
300.9 3.1
21.2 0.6
(7.0%)

24.3 0.3
16.2 0.2
(66.7%)
23.9 0.6
5.0 0.1
(20.9%)

26.7 0.7
18.8 0.4
(70.4%)
107.2 1.0
9.1 0.3
(8.5%)

16.2 0.4
11.1 0.2
(68.5%)
8.1 0.1
1.0 0.0
(12.3%)

450.2 6.9
292.3 4.3
(64.9%)
1003.5 8.5
270.6 3.7
(27.0%)

Total phenolics
(mg C L1)

1163.7 20.2
470.5 13.7
(40.4%)
2603.8 20.2
1150.5 33.7
(44.2%)

TC: total catechins, the sum of individual catechins.

Control: 3 mL tea infusion stock mixed with 2 mL water.
c
Data in the blanket were distribution ratios of tea catechins or TP in the skimmed serum fractions: distribution ratio = (the concentration of catechins or TP in control the
concentration of catechins or TP in the skimmed serum) / the concentration of catechins or TP in control 100%. The distribution ratios of catechins or TP in control were set as
100%.
b

in the BTPmilk system while casein micelles incorporated most of tea

catechins in the GTPmilk system. 35.1% and 73% of TC were bound to
the casein micelles for the BTP and GTPmilk systems (Table 1),
which were analogous to the contributions of tea catechins (16% and
80%) to the masking effects of caseins on the antioxidant activities of
black tea and green tea (Arts et al., 2002). It may infer that caseins are
the major cause for the masking effect of milk on the antioxidant potentials of TP. In our study, milk fat only incorporated trace levels of TP,
which is different from the result that milk fat content affected the
masking effect of milk on the antioxidant potentials of green tea and
black tea (Langley-Evans, 2000a). The different ratios of caseins/whey
proteins are a possible explanation for the varying masking effects of
milk with different fat content but similar level of protein.

3.2. FTIR spectra of the skimmed milk and TPmilk samples

Table 2 and Fig. 1 show the secondary structures of milk proteins
with or without interaction with TP analyzed by FTIR. As shown in
Fig. 1, the skimmed milk proteins contained 7.7% -helix (1653 cm1),
15.6% inter -sheet (1614 and 1623 cm1) and 9.0% intra -sheet
(1633 cm1), 23.7% turn structure (1680 cm1), 3.6% -antiparallel
(1693 cm1), 25.8% large loop (1660 cm1) and 14.6% random coil
(1643 cm1), which was in agreement with the spectroscopic studies
of caseins previously reported (Alaimo, Farrell, & Germann, 1999; Byler
& Susi, 1986). Upon TP interaction, -helix and intra -sheet increased
from 7.7% and 9.0% to 20.7% and 28.1% for BTP as well as 18.9% and
26.4% for GTP, whereas inter -sheet decreased from 15.6% to 5.0% and
3.4% accordingly (Fig. 1). For the irregular segments in milk proteins,
random coil and the large loop decreased from 14.6% and 25.8% to 0.9%
and 8.8% for BTP as well as 0.3% and 15.6% for GTP, but turn structure
increased from 23.7% to 32.4% and 30.7% respectively upon interaction
with BTP and GTP (Fig. 1 and Table 2).
Table 2
Secondary structure analysis of the skimmed milk, BTP and GTPmilk samples.
Amide I
components

Skimmed milk

BTPmilk sample

GTPmilk sample

16241610 cm1
(inter--sheet)
16451625 cm1
(intra--sheet)
16481641 cm1
(random coil)
1652 2 cm1
(-helix)
1658 2 cm1
(large loop)
16851660 cm1
(turn)
17001690 cm1
(-anti)

5.0%, 1614 cm1;

10.6%, 1623 cm1
9.0%, 1633 cm1

5.0%, 1614 cm1

3.4%, 1616 cm1

28.1%, 1633 cm1

26.4%, 1633 cm1

14.6%, 1643 cm1

0.9%, 1643 cm1

0.3%, 1643 cm1

7.7%, 1653 cm1

20.7%, 1651 cm1

18.9%, 1651 cm1

25.8%, 1660 cm1

8.8%, 1660 cm1

15.6%, 1660 cm1

23.7%, 1680 cm1

12.6%,1668 cm1;
19.8%,1680 cm1
4.1%, 1693 cm1

27.6%, 1670 cm1;

3.1%, 1680 cm1
4.7%, 1691 cm1

3.6%, 1693 cm1

Fig. 1. Second derivative resolution enhancement and curve-tted amide I region

(17001600 cm1) for the skimmed milk (A), BTP and GTPmilk samples (B and C).

453

J. Ye et al. / Food Research International 53 (2013) 449455

Fig. 2 shows the hydrophobic interactions between TP and milk

proteins in the FTIR spectra region of 30002800 cm1 assigned to
CH2 bands. The absorption bands of milk proteins located at
2923.74 and 2852.69 cm1 were attributed to antisymmetric and
symmetric CH2 stretching vibrations. Upon interaction with TP,
more reduction in the absorption intensities of the BTPmilk sample
was observed compared to the GTPmilk sample (Fig. 2). The weakening of antisymmetric and symmetric CH2 stretching vibrations suggested the presence of hydrophobic interactions between TP and milk
proteins, and the hydrophobic interactions between BTP and milk
proteins were more intense.
Caseins belong to an unstructured protein family with unique unfolded structure under native conditions, due to the high content of
proline residues (Alaimo et al., 1999; Hasni et al., 2011). Proline is
known to disrupt the regular structures of proteins, such as -helix
and inter -sheet, but excels at forming turns and hydrophobic regions (Alaimo et al., 1999), which is consistent with the above FTIR
results. Upon interaction with TP, the proline residues on proteins
interacted with TP due to the strong afnity of polypenols for proline
residues (Siebert, Troukhanova, & Lynn, 1996), hence the irregular
structures of proteins like random coil and the large loop, as well as
intra -sheet, were interrupted. The released segments of proteins
were partially folded into -helix or intra -sheet driven by hydrogen
bond, but the solvent-exposed part tends to form turns more
according to Alaimo et al. (1999). That is why major increases were
observed in -helix, intra -sheet and turn structures after interaction with TP. More hydrophobic large loop was interrupted by BTP
than GTP (Fig. 1), which is consistent with the stronger hydrophobic
interaction between BTP and milk proteins as shown in Fig. 2. The hydrophobic interactions between TP and milk proteins are due to the
accommodation of aromatic rings of TP on the hydrophobic regions
of proteins (Spencer et al., 1988). It may infer that the highly polymerized polyphenols in black tea with more conjugated and
galloylated structures are more apt to interact with the hydrophobic
regions of milk proteins.
3.3. UVvis spectra study

2923.74

The UV absorptions of black tea infusion and green tea infusion are
attributed to electronic transitions between -type molecular orbitals
(MOs) in TP molecules (Anouar, Gierschner, Duroux, & Trouillas,
2012), which mainly occurs to aromatic rings. The substituent groups
(other than H) or hydrogen bond attached to aromatic rings alter the
maximum wavelength or the intensity of UV absorption. Fig. 3 shows

the UVvis spectra of BTP and GTPmilk systems at low concentration of milk proteins. With an increasing addition of milk, the intensity of UV absorption for BTP and GTPmilk system increased
gradually, and no obvious shift of absorption maximum wavelength
was observed (Fig. 3). Since the UV absorption of milk was already
subtracted, the changes in UV spectra were attributed to the interaction between TP and milk proteins. An increase in UV absorption intensity of TP indicated that hydrogen bonding might occur between
the phenolic hydroxyl of TP and the amide group of milk proteins
(Haslam, Williamson, Baxter, & Charlton, 1999), because hydrogen
bonding increased the intensity of electron clouds on the aromatic
rings of TP and lowered the transition energy, leading to a
hyperchromic effect. No obvious shift observed in Fig. 3 meant no
new conjugated double bonds or extending conjugations are generated to TP molecules upon interaction with a low concentration of milk
proteins.

3.4. Fluorescence properties of the TPmilk systems and the correlation

with reductions in ORAC values
Fig. 4 shows the uorescence emission spectra of milk proteins
quenched by BTP and GTP. No spectral shift was observed in Fig. 4, indicating that the interactions with BTP and GTP have no inuence on
the environment of uorophores in proteins. The uorescence intensities of milk proteins decreased with increasing additions of BTP and
GTP, and the corresponding quenching ratio% was shown in Fig. 4A.
BTP exhibited a similar quenching ability to GTP (Fig. 4A), but the
variation in the masking% of the ORAC values of the BTPmilk system
was different from that of the GTPmilk system (Table 3). A signicant correlation (r = 0.725) between the uorescence quenching
ratio and masking% of the ORAC values was observed for the BTP
milk system in comparison with no signicant correlation for the
GTPmilk system (Table 3). As the addition of black tea increased
from 50 to 200 L, the masking% of the ORAC value of the BTPmilk
system increased from 2.3% to 10.8%. As the addition of black tea further increased, the masking% of the ORAC value declined slightly, and
then sharply increased up to the maximum of 54.3%. Similar tendency
was found in the GTPmilk system but only 17.3% of the ORAC value
diminished at the highest level of GTP (Table 3).
BTP exerted a similar uorescence quenching ability to GTP at the
same addition of tea infusion despite the lower concentration of total
phenolics than that of green tea infusion (Table 1). This might be due
to the same accessibility of Trp and Tyr residues to the excessive BTP
and GTP in tea infusions. However, the masking% in the ORAC values
of the BTP and GTPmilk systems did not change accordingly with
1.2

GTP-milk sample

3000

2960

2852.69

1.0

2920

Absorbance

2851.90

2880

BTP
GTP

a
f

2852.98

2917.80
2921.93

Absorbance

Skimmed milk

BTP-milk sample

0.8

a
0.6

2840

2800

Wavenumber (cm-1)
Fig. 2. FTIR spectra in the region of 30002800 cm1 for the skimmed milk, BTP and
GTPmilk samples. The contributions of the CH2 stretching vibrations of BTP and GTP
in the region 30002800 cm1 were subtracted.

0.4
260

270

280

290

Wavelength (nm)
Fig. 3. UV absorption spectra of BTP and GTP upon interaction with whole milk. Tea
infusion stock (100 L) was mixed with 0, 20, 40, 80, 160 and 320 L (af) of 100-fold
diluted whole milk, and then made up to 10 mL.

454

J. Ye et al. / Food Research International 53 (2013) 449455

100

BTP
GTP

Quenching ratio (%)

Fluorecense intensity (a.u)

2500

2000

1500

affect the afnities of tea catechins for casein micelles in the GTPmilk
system but no obvious impact for the BTPmilk system. A gallate
group and the cis-form enhance the catechincasein interaction, while
a pyrogallol group weakens the interaction. The interactions with TP altered the secondary structures of milk proteins by interrupting the native irregular structures of random coil and the large loop and
increasing -helix, intra -sheet and turn structures. UVvis spectra
showed no new conjugated double bonds or extending conjugations
that occur to TP molecules upon interaction with a low concentration
of milk proteins. The uorescence intensities of milk proteins decreased
gradually with increasing additions of BTP and GTP. Despite the similar
uorescence quenching abilities of BTP and GTP, the variations in the
masking% of the ORAC values for the BTP and GTPmilk system
were different. Thus, uorescence quenching may not be used as
the unique method to interpret the interactions between polyphenols
and proteins.

60
40
BTP
GTP

20
0

1000

A'

80

100 200 300 400 500 600

Tea infusion (uL)

500

0
300

320

340

360

380

400

420

440

Wavelength (nm)
Fig. 4. Fluorescence emission spectra of milk proteins upon interaction with BTP and GTP.
Whole milk (1200 L) was mixed with 0 (a), 50, 100, 150, 200, 250, 300, 400, 500 and
600 L (bj) tea infusion stock respectively, and then made up to 2000 L. Inset (A):
the corresponding quenching ratio. Quenching ratio % = (F0 F) / F0 100%.

References

the corresponding quenching ratios, suggesting more interaction regions for TP on proteins besides the Trp/Tyr sites. Two stages were
found in the variation of masking% for the BTP and GTPmilk systems, suggesting that two types of polyphenolprotein interactions
might occur as the concentration of TP gradually increased. At low
concentration of TP, TP is easily accommodated onto the relative hydrophilic surface of casein micelles. As the concentration of TP further
increased, the TP molecules penetrate into the internal part of casein
micelles and get more access to the hydrophobic regions. Since BTP
contains plenty of conjugated structures with aromatic rings and quinones (Balentine, Wiseman, & Bouwens, 1997), BTP is likely to have
stronger hydrophobic interactions with the internal part of casein micelles compared to GTP, which is consistent with the result in Table 3
that 54.3% and 17.3% of the ORAC values were masked at the highest
levels of BTP and GTP. Therefore only using uorescence quenching to
interpret polyphenolprotein interaction may not be sufcient to reect various types of interactions between polyphenols and proteins.
4. Conclusions
In the BTP and GTPmilk systems, most BTP and GTP were combined with milk proteins. Casein micelles are more likely to bind highly
polymerized polyphenols, while whey proteins are preferential to bind
and transport smaller molecules. The behaviors of tea catechins in the
BTP and GTPmilk systems were different. The structures of catechins

Alaimo, M. H., Farrell, H. M., & Germann, M. W. (1999). Conformational analysis of the
hydrophobic peptide s1-casein(136196). Biochimica et Biophysica Acta-Protein
Structure and Molecular Enzymology, 1431, 410420.
Anema, S. G., & Klostermeyer, H. (1997). Heat-induced, pH-dependent dissociation of
casein micelles on heating reconstituted skim milk at temperatures below
100 C. Journal of Agricultural and Food Chemistry, 45, 11081115.
Anouar, E., Gierschner, J., Duroux, J. L., & Trouillas, P. (2012). UV/Visible spectra of
natural polyphenols: A time-dependent density functional theory study. Food
Chemistry, 131, 7989.
Arts, M. J. T. J., Haenen, G. R. M. M., Wilms, L. C., Beetstra, S. A. J. N., Heijnen, C. G. M.,
Voss, H. P., et al. (2002). Interactions between avonoids and proteins: Effect on
the total antioxidant capacity. Journal of Agricultural and Food Chemistry, 50,
11841187.
Balentine, D. A., Wiseman, S. A., & Bouwens, L. C. M. (1997). The chemistry of tea avonoids.
Critical Reviews in Food Science and Nutrition, 37, 693704.
Baxter, N. J., Lilley, T. H., Haslam, E., & Williamson, M. P. (1997). Multiple interactions
between polyphenols and a salivary proline-rich protein repeat result in complexation and precipitation. Biochemistry, 36, 55665577.
Benzie, I. F. F., & Szeto, Y. T. (1999). Total antioxidant capacity of teas by the ferric
reducing/antioxidant power assay. Journal of Agricultural and Food Chemistry, 47,
633636.
Bohin, M. C., Vincken, J. P., van der Hijden, H. T. W. M., & Gruppen, H. (2012). Efcacy of
food proteins as carriers for avonoids. Journal of Agricultural and Food Chemistry,
60, 41364143.
Bourassa, P., Kanakis, C. D., Tarantilis, P., Pollissiou, M. G., & Tajmir-Riahi, H. A. (2010).
Resveratrol, genistein, and curcumin bind bovine serum albumin. The Journal of
Physical Chemistry B, 114, 33483354.
Byler, D. M., & Susi, H. (1986). Examination of the secondary structure of proteins by
deconvolved ftir spectra. Biopolymers, 25, 469487.
Corredig, M., & Dalgleish, D. G. (1996). Effect of temperature and pH on the interactions of
whey proteins with casein micelles in skim milk. Food Research International, 29, 4955.
Curley, D. M., Kumosinski, T. F., Unruh, J. J., & Farrell, H. M. (1998). Changes in the
secondary structure of bovine casein by Fourier transform infrared spectroscopy:
Effects of calcium and temperature. Journal of Dairy Science, 81, 31543162.
Dubeau, S., Samson, G., & Tajmir-Riahi, H. A. (2010). Dual effect of milk on the antioxidant capacity of green, Darjeeling, and English breakfast teas. Food Chemistry, 122,
539545.

Table 3
The ORAC values of BTP and GTPmilk systems (M trolox).
Tea infusion (L)

0 (pure milk)
50
100
150
200
250
300
400
500
600
Corr. Coeff. c
a
b
c
d

BTP

529.2 6.3
652.3 8.2
1101.7 11.2
1814.5 23.3
2459.6 32.7
3125.2 41.5
4172.7 52.9
5674.1 38.6
7115.6 57.7

BTPmilk system

GTPmilk system

Ca. valuea

Ex. Valuea

9200.4
9323.5
9772.9
10,485.7
11,130.8
11,796.4
12,843.9
14,345.3
15,786.8

8671.2
8987.7
8816.4
9020.9
9355.6
10,108.6
10,850.3
10,646.1
7037.3
7216.0

69.5
33.3
50.2
84.2
111.8
46.7
84.3
76.4
69.3
76.6

Masking%b

GTP

0%
2.3%
5.4%
7.7%
10.8%
9.2%
8.0%
17.1%
50.9%
54.3%
0.725d

1390.4
1961.1
2834.6
3855.0
4508.9
5350.1
7009.9
8465.5
9784.5

16.2
21.3
13.6
26.3
39.8
56.7
66.4
96.2
106.7

Ca. valuea

Ex. valuea

10,000.2
10,570.9
11,444.4
12,464.8
13,118.7
13,959.9
15,619.7
17,075.3
18,394.3

8609.8
8570.4
8743.3
9339.1
9948.9
10,664.7
12,289.8
13,682.9
14,612.2
15,220.6

Ca. value: Calculated value; Ex. Value: Experimental value. Calculated ORAC value = the ORAC value of pure milk + the ORAC value of TP.
Masking% = (Ca. value ex. value) / Ca. value 100%; the masking% of pure milk was set as 0%.
Corr. Coeff.: Pearson's correlation coefcient between quenching ratio (%) and masking percentage (%).
Indicated the correlation is signicant at the 0.05 level. NS indicated no signicant difference.

Masking%b
73.5
40. 1
52.3
100.2
91.2
79.9
81.4
103.5
155.2
131.6

0%
14.3%
17.3%
18.4%
20.2%
18.7%
12.0%
12.4%
14.4%
17.3%
NS

J. Ye et al. / Food Research International 53 (2013) 449455

Dufresne, C., & Farnworth, E. (2000). Tea, Kombucha, and health: A review. Food
Research International, 33, 409421.
Farrell, H. M., Wickham, E. D., Unruh, J. J., Qi, P. X., & Hoagland, P. D. (2001). Secondary
structural studies of bovine caseins: Temperature dependence of -casein structure as analyzed by circular dichroism and FTIR spectroscopy and correlation
with micellization. Food Hydrocolloids, 15, 341354.
Ferruzzi, M. G., & Green, R. J. (2006). Analysis of catechins from milktea beverages by
enzyme assisted extraction followed by high performance liquid chromatography.
Food Chemistry, 99, 484491.
Garcia-Risco, M. R., Ramos, M., & Lopez-Fandino, R. (2002). Modications in milk
proteins induced by heat treatment and homogenization and their inuence on
susceptibility to proteolysis. International Dairy Journal, 12, 679688.
Goto, T., Yoshida, Y., Kiso, M., & Nagashima, H. (1996). Simultaneous analysis of individual
catechins and caffeine in green tea. Journal of Chromatography. A, 749, 295299.
Gupta, S., Saha, B., & Giri, A. K. (2002). Comparative antimutagenic and anticlastogenic
effects of green tea and black tea: A review. Mutation Research, Reviews in Mutation
Research, 512, 3765.
Haslam, E., Williamson, M. P., Baxter, N. J., & Charlton, A. J. (1999). Astringency and
polyphenol protein interactions. Recent Advances in Phytochemistry, 33, 289.
Hasni, I., Bourassa, P., Hamdani, S., Samson, G., Carpentier, R., & Tajmir-Riahi, H. A.
(2011). Interaction of milk - and -caseins with tea polyphenols. Food Chemistry,
126, 630639.
Hemar, Y., Gerbeaud, M., Oliver, C. M., & Augustin, M. A. (2011). Investigation into the
interaction between resveratrol and whey proteins using uorescence spectroscopy.
International Journal of Food Science and Technology, 46, 21372144.
Huang, D. J., Ou, B. X., Hampsch-Woodill, M., Flanagan, J. A., & Prior, R. L. (2002).
High-throughput assay of oxygen radical absorbance capacity (ORAC) using a
multichannel liquid handling system coupled with a microplate uorescence reader
in 96-well format. Journal of Agricultural and Food Chemistry, 50, 44374444.
Kanakis, C. D., Hasni, I., Bourassa, P., Tarantilis, P. A., Polissiou, M. G., & Tajmir-Riahi,
H. A. (2011). Milk -lactoglobulin complexes with tea polyphenols. Food Chemistry, 127, 10461055.
Komes, D., Horzic, D., Belscak, A., Ganic, K. K., & Vulic, I. (2010). Green tea preparation
and its inuence on the content of bioactive compounds. Food Research International,
43, 167176.
Kyle, J. A., Morrice, P. C., McNeill, G., & Duthie, G. G. (2007). Effects of infusion time and
addition of milk on content and absorption of polyphenols from black tea. Journal
of Agricultural and Food Chemistry, 55, 48894894.
Lakowicz, J. R. (1999). Principles of uorescence spectroscopy (3rd ed.)New York: Kluwer
Academic/Plenum.
Langley-Evans, S. C. (2000a). Antioxidant potential of green and black tea determined
using the ferric reducing power (FRAP) assay. International Journal of Food Sciences
and Nutrition, 51, 181188.

455

Langley-Evans, S. C. (2000b). Consumption of black tea elicits an increase in plasma antioxidant potential in humans. International Journal of Food Sciences and Nutrition,
51, 309315.
Leenen, R., Roodenburg, A. J. C., Tijburg, L. B. M., & Wiseman, S. A. (2000). A single dose
of tea with or without milk increases plasma antioxidant activity in humans. European Journal of Clinical Nutrition, 54, 8792.
Livney, Y. D. (2010). Milk proteins as vehicles for bioactives. Current Opinion in Colloid
& interface Science, 15, 7383.
Lorenz, M., Jochmann, N., von Krosigk, A., Martus, P., Baumann, G., Stangl, K., et al.
(2007). Addition of milk prevents vascular protective effects of tea. European
Heart Journal, 28, 219223.
Oteiza, P. I., Erlejman, A. G., Verstraeten, S. V., Keen, C. L., & Fraga, C. G. (2005). Flavonoid
membrane interactions: A protective role of avonoids at the membrane surface?
Clinical and Developmental Immunology, 12, 1925.
Richelle, M., Tavazzi, I., & Offord, E. (2001). Comparison of the antioxidant activity of
commonly consumed polyphenolic beverages (coffee, cocoa, and tea) prepared
per cup serving. Journal of Agricultural and Food Chemistry, 49, 34383442.
Sahu, A., Kasoju, N., & Bora, U. (2008). Fluorescence study of the curcumincasein
micelle complexation and its application as a drug nanocarrier to cancer cells.
Biomacromolecules, 9, 29052912.
Serani, M., Ghiselli, A., & FerroLuzzi, A. (1996). In vivo antioxidant effect of green and
black tea in man. European Journal of Clinical Nutrition, 50, 2832.
Sharma, V., Kumar, H. V., & Rao, L. J. M. (2008). Inuence of milk and sugar on antioxidant
potential of black tea. Food Research International, 41, 124129.
Siebert, K. J., Troukhanova, N. V., & Lynn, P. Y. (1996). Nature of polyphenolprotein
interactions. Journal of Agricultural and Food Chemistry, 44, 8085.
Spencer, C. M., Cai, Y., Martin, R., Gaffney, S. H., Goulding, P. N., Magnolato, D., et al. (1988).
Polyphenol complexation Some thoughts and observations. Phytochemistry, 27,
23972409.
Staszewski, M. V., Jara, F. L., Ruiz, A. L. T. G., Jagus, R. J., Carvalho, J. E., & Pilosof, A. M. R.
(2012). Nanocomplex formation between -lactoglobulin or caseinomacropeptide
and green tea polyphenols: Impact on protein gelation and polyphenols
antiproliferative activity. Journal of Functional Foods, 4, 800809.
Xiao, J. B., Mao, F. F., Yang, F., Zhao, Y. L., Zhang, C., & Yamamoto, K. (2011). Interaction
of dietary polyphenols with bovine milk proteins: molecular structureafnity relationship and inuencing bioactivity aspects. Molecular Nutrition & Food Research,
55, 16371645.
Zulueta, A., Esteve, M. J., & Frigola, A. (2009a). ORAC and TEAC assays comparison to
measure the antioxidant capacity of food products. Food Chemistry, 114,
310316.
Zulueta, A., Maurizi, A., Frigola, A., Esteve, M. J., Coli, R., & Burini, G. (2009b). Antioxidant
capacity of cow milk, whey and deproteinized milk. International Dairy Journal, 19,
380385.