LWT - Food Science and Technology 64 (2015) 706e712

Contents lists available at ScienceDirect

LWT - Food Science and Technology

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

The improved thermal stability of anthocyanins at pH 5.0 by gum

arabic
Yongguang Guan, Qixin Zhong*
Department of Food Science and Technology, The University of Tennessee, Knoxville, TN 37996, USA

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 14 March 2015
Received in revised form
2 June 2015
Accepted 6 June 2015
Available online 17 June 2015

Gum arabic (GA) was studied to improve the thermal stability of anthocyanins. Solutions with 0.51 mg/
mL anthocyanins at pH 5.0 were heated at 80 and 126
C up to 80 min with and without 10 mg/mL GA.
The half-life of thermal degradation of anthocyanins at 80 and 126
C was increased by 2.0 and 1.8 times,
respectively, after adding GA. The residual concentration of anthocyanins with GA was 1.02 and 1.35
times higher than that without GA after 30-min heating at 80 and 126
C, respectively. The DPPH and
ABTS free-radical scavenging capacities and ferric reducing power after heating anthocyanin solutions
with GA at 126
C for 30 min were signicantly higher than those without GA. Fluorescence spectroscopy
data suggested the formation of complexes between anthocyanins and GA. Zeta-potential data suggested
the formation of complexes by hydrophobic attraction. Stable particle size and zeta-potential of GA with
and without anthocyanins were observed after heating. This study indicated that the assembled GA
nanostructures signicantly reduced the thermal degradation of anthocyanins at pH 5.0.
2015 Elsevier Ltd. All rights reserved.

Keywords:
Anthocyanins
Gum arabic
Complex formation
Hydrophobic attraction
Heat stability

1. Introduction
Anthocyanins are water-soluble pigments delivering a clear
color that changes from salmon pink, red, violet, to dark blue as pH
increases (Cavalcanti, Santos, & Meireles, 2011). Anthocyanins are
present at high concentrations in various food products (Santos &
Meireles, 2009) and have been studied for their potential biological activities improving human health such as reduced risks of
chronic diseases (Cavalcanti et al., 2011) and coronary heart disease
(Colantuoni, Bertuglia, Magistretti, & Donato, 1991). Anthocyanins
have also shown good antioxidant (Cavalcanti et al., 2011), antiinammatory (J. Wang & Mazza, 2002), antimutagenic (Peterson
& Dwyer, 1998), and chemo preventive activities (Zhao, Giusti,
Malik, Moyer, & Magnuson, 2004). Despite these ndings, anthocyanins have not been used as therapeutic and health-promoting
agents (Cavalcanti et al., 2011).
One of the major challenges limiting the application of anthocyanins is their instability during storage and processing. The
degradation of anthocyanins involves the hydrolysis of the double
bond of ring C of avylium cation via intramolecular charge
transfer, which is more signicant above pH 4.5 (Brouillard &
* Corresponding author. Department of Food Science and Technology, The University of Tennessee, 2510 River Drive, Knoxville, TN 37996, USA. Tel.: 1 865 974
6196; fax: 1 865 974 7332.
E-mail address: qzhong@utk.edu (Q. Zhong).
http://dx.doi.org/10.1016/j.lwt.2015.06.018
0023-6438/ 2015 Elsevier Ltd. All rights reserved.

Delaporte, 1977). The degradation of anthocyanins during thermal

processing and storage can be accelerated by oxygen (OdriozolaSerrano, Soliva-Fortuny, & Martn-Belloso, 2010) and light (Maier,
Fromm, Schieber, Kammerer, & Carle, 2009).
Approaches studied for stabilizing anthocyanins include
encapsulation, self-association, co-pigmentation, and adoption of
metallic ions (Cavalcanti et al., 2011). Among these methods,
encapsulation has been a rapidly expanding area because immobilization of bioactive compounds in particles can protect them
against degradation (Oidtmann et al., 2012). Technologies of
encapsulating anthocyanins have been studied for spray drying
(Main, Clydesdale, & Francis, 1978), lyophilization, thermal gelation, and ionic gelation (Cavalcanti et al., 2011). Encapsulation
materials have been studied for maltodextrin (Ersus & Yurdagel,
2007), glucan (Xiong, Melton, Easteal, & Siew, 2006), pectin, sodium alginate, gurdlan (Cavalcanti et al., 2011) and gum arabic (GA)
(Valduga, Lima, Prado, Padilha, & Treichel, 2008), with some studies
reporting the improved heat stability of anthocyanins after
encapsulation. Recently, we showed that mannoproteins extracted
from the yeast cell wall formed complexes with and improved
thermal stability of anthocyanins at pH 7.0 (Wu, Guan, & Zhong,
2015). For clear beverage products, water-soluble biopolymers
giving a low viscosity are to be studied.
GA is a commonly-used ingredient with a high water solubility
and low solution viscosity (Gomes et al., 2010). GA is known for its

Y. Guan, Q. Zhong / LWT - Food Science and Technology 64 (2015) 706e712

ability to form stable emulsions over a wide range of acidity and

ionic conditions (Gomes et al., 2010; Jayme, Dunstan, & Gee, 1999).
GA is a glycoprotein with polysaccharide chains composed of six
types of monomers (galactopyranose, arabinopyranose, arabinofuranose, rhamnopyranose, glucuropyranosyluronic acid and 4-Omethylglucuro-pyranosyluronicacid) and a small fraction of protein
(Jayme et al., 1999). The periphery of GA has a protein:carbohydrate
molar ratio of about 1:40, while this ratio is 1:11 at the galactan
core (Islam, Phillips, Sljivo, Snowden, & Williams, 1997). Both acidic
amino acid residues of the protein portion and carbonyl groups of
carbohydrates contribute to negative surface charges of GA (Islam
et al., 1997). Besides electrostatic repulsion, the steric hindrance
originating from polysaccharides after the polypeptide moiety of
GA adsorbing to hydrophobic surfaces contributes to the excellent
stabilization function of GA in various colloidal systems (Buffo,
Reineccius, & Oehlert, 2001; Williams, Gold, Holoman, Ehrman, &
Wilson Jr, 2006). In our recent study, GA was observed to form
complexes with water-soluble pigment norbixin, and the complexes improved the thermal stability of norbixin at pH 3.0e5.0
(Guan & Zhong, 2014). Currently, there is no study about the heat
stability of anthocyanins above pH 4.5 when they are co-dissolved
with GA.
The rst objective of the present study was to evaluate thermal
degradation of anthocyanins at 80 and 126
C as improved by GA at
pH 5.0. These two temperatures were chosen because of the relevance to pasteurization (Berk, 2008) and sterilization (targeting
spores of Bacillus stearothermophilus ATCC 7953) (Marques, Narita,
Costa, & Rezende, 2010). The pH 5.0 was chosen because anthocyanin solutions are degraded to be colorless after thermal treatments at 70, 80 and 90
C (Xu, Liu, Yan, Yuan, & Gao, 2015). The
second objective was to understand physicochemical properties of
anthocyanin-GA mixture solutions before and after heating at 80
and 126
C for 30 min. The third objective was to evaluate impacts
of thermal processing on antioxidation properties of anthocyanins
as improved by GA.
2. Materials and methods
2.1. Chemicals
The anthocyanin sample (ColorFruit Violet 100WS, 540 mg/mL)
was obtained from Chr. Hansen (Hrsholm, Denmark). GA, 2,2Diphenyl-1-picrylhydrazyl (DPPH), potassium ferricyanide, and
ferric chloride were purchased from Fisher Scientic (Pittsburgh,
PA). 2,20 -Azino-bis(3-ethylbenzothiazoline-6-sulfonic acid) diammonium salt (ABTS), potassium persulfate, and trichloroacetic
acid were purchased from SigmaeAldrich Corp. (St. Louis, MO).
2.2. Preparation of samples
Anthocyanins were dissolved in distilled water, followed by
dissolving GA powder to a nal concentration of 0.51 mg/mL
anthocyanin and 10 mg/mL GA. After stirring for 30 min, the
mixture was adjusted to pH 5.0 using 1 mol/L HCl and 1 mol/L
NaOH. Solutions with and without GA were heated at 80 and 126
C
for up to 80 min to determine degradation kinetics, while those
heated for 30 min were used to characterize other properties. All
samples were placed at 4
C in dark before further tests within 48 h.
2.3. Color and concentration of anthocyanins
The effects of GA on the color stability of anthocyanin solutions
were evaluated for L*, a*, and b* values. Five milliliters of a sample
was measured at room temperature (21
C) using a MiniScan XE
Plus Hunter colorimeter (Hunter Associates Laboratory, Inc., Reston,

707

VA), following the protocol in a previous study (Zhang & Zhong,

2013).
The absorbance of samples before and after heating was
measured at 570 nm to determine the anthocyanin concentration
using an Evolution 201 UVeVis spectrophotometer (Thermo Scientic, Waltham, MA). Two calibration curves were established for
samples with and without GA using standard solutions with
various anthocyanin concentrations with and without 10 mg/mL
GA adjusted to pH 5.0.
2.4. Thermal degradation kinetics of anthocyanins
In this set of samples, the above anthocyanin solutions at pH 5.0
were heated at 80 or 126
C for 0, 10, 20, 30, 40, 60 and 80 min. After
cooling to room temperature (21
C) in a water bath, the absorbance at 570 nm was tested, and the determined anthocyanin
concentrations were t to Eq. (1) based on the rst order degra
lu, 2007; W.-D. Wang &
dation kinetics (Krca, Ozkan,
& Cemerog
Xu, 2007). The half-life (t1=2 ), the duration corresponding to
anthocyanin concentration reducing to one-half of the starting
concentration, was calculated according to Eq. (2) (Gris, Ferreira,
Falc~
ao, & Bordignon-Luiz, 2007).

ct co ekt

(1)

t1=2 ln0:5$k1

(2)

where co and ct are anthocyanin concentrations before and after

heating at 80 or 126
C for a duration of t (min), and k is the
degradation rate constant at 80 or 126
C (min1).
2.5. Fluorescence spectroscopy
Fluorescence spectra of GA-anthocyanin solutions were determined according to the method in the literature (Zhang & Zhong,
2012) with a slight modication. The instrument was a model RF1501 spectrouorometer (Shimadzu Corp., Tokyo, Japan). The
excitation wavelength was set at 330 nm, and the emission spectra
were recorded from 220 to 600 nm with background uorescence
calibrated using distilled water.
2.6. Particle size and zeta-potential measurements
Samples were diluted 10 times using distilled water and
adjusted to pH 5.0. Samples were analyzed for hydrodynamic diameters using a Delsa Nano-Zeta Potential and Submicron Particle Size Analyzer (Beckman Coulter Inc., Brea, CA) at room
temperature (21
C). Samples at pH 5.0 without dilution were
tested for zeta-potential (Zetasizer Nano ZS, Malvern Instruments
Ltd, Worcestershire, UK).
2.7. Atomic force microscopy (AFM)
AFM assay of structures before and after heating followed the
protocol in our previous study (Pan & Zhong, 2013). Samples were
diluted 100 times in distilled water and readjusted to pH 5.0. Five
microliters of the diluted sample was dropped onto a freshly-cleaved
mica sheet with an area of ~1.8 cm2 and dried at room temperature
(21
C) overnight in a fume hood. Topography images were collected
at the tapping mode using a nano-probe cantilever tip (Bruker Corp.,
Santa Barbara, CA) driven at a frequency from 50 to 100 kHz on a
Multimode VIII microscope (Bruker Corporation, Billerica, MA). Images were analyzed using the AMF instrument software (Nanoscope
Analysis version 1.50, Bruker Corporation, Billerica, MA).

708

Y. Guan, Q. Zhong / LWT - Food Science and Technology 64 (2015) 706e712

2.8. Determination of antioxidant capacity

2.8.1. DPPH radical-scavenging capacity
DPPH assay was carried out according to the method reported in
rillon,
the literature (Dudonne, Vitrac, Coutiere, Woillez, & Me
2009) with slight modications. DPPH radicals (DPPH$) have the
maximum absorption at about 517 nm that is reduced by an antioxidant. The fresh DPPH$ solution was prepared daily at
0.125 mmol/L in ethanol. After mixing 1 mL of a sample solution
and 1 mL of the DPPH$ solution, the mixture was incubated at room
temperature (21
C) in dark for 20 min before measuring absorbance at 517 nm (model Evolution 201, Thermo Scientic, Waltham,
MA). The blank sample was prepared by substituting distilled water
for anthocyanin samples. The DPPH radical-scavenging activity was
calculated using Eq. (3):

Radical scavenging activity % 1  AE =AB  100%

(3)

where AE and AB are the respective absorbance value of a sample

and the blank.
2.8.2. ABTS radical-scavenging capacity
ABTS assay was carried out according to a literature method (Re
et al., 1999) with some modication. In the assay, cationic ABTS
radicals (ABTS$) are reduced by an antioxidant, resulting in the
reduced absorbance at 734 nm. The fresh ABTS stock solution
(7 mmol/L in distilled water) was mixed with 2.45 mmol/L potassium persulfate and placed at 21
C in dark for 16 h to generate
ABTS$. The ABTS$ stock solution was diluted 10 times in distilled
water as a working solution. The 0.1 mL of a sample solution was
mixed with 3 mL of the ABTS$ working solution. After incubation
at 30
C in dark for 10 min, the absorbance was measured at 734 nm
using the above UVeVis spectrophotometer. Distilled water was
used to prepare a blank by substituting a sample in the assay. The
ABTS$ scavenging activity (%) was calculated using the above Eq.
(3).
2.8.3. Reducing power
The reducing power of samples was determined according to a
method reported previously (Lertittikul, Benjakul, & Tanaka, 2007),
with some modication. All solutions were prepared in distilled
water. One half mL of a sample was mixed with 0.5 mL of 0.2 mol/L
sodium phosphate buffer (pH 6.6) and 0.5 mL of 10 mg/mL potassium ferricyanide. The mixture was incubated at 50
C for 20 min
before adding 1 mL of 100 mg/mL trichloroacetic acid, 1.5 mL of
distilled water, and 0.3 mL of 1 mg/mL FeCl3. The blank was prepared as above, except that the 10 mg/mL potassium ferricyanide

solution was replaced by distilled water. The absorbance at 700 nm

was measured and reported as the reducing power after subtraction by the absorbance of the blank at the same wavelength.
2.9. Statistical analysis
All experiments were carried out in triplicate. The mean and
standard error of each treatment were calculated from replicates.
Analysis of variance (ANOVA) was carried out to determine any
signicant differences between treatments (P < 0.05).
3. Results and discussion
3.1. Color stability of anthocyanin solutions
GA signicantly improved the color stability of anthocyanin
solution after heating at 80 and 126
C for 30 min, based on both
visual appearance (Fig. 1) and quantitative color parameters
(Table 1). The b* value signicantly increased, while a* value
signicantly decreased after heating. Changes of a* and b* values
after heating became less after GA addition, which indicated the
protection of anthocyanins by GA. Lightness (L* value) of solutions
without GA increased signicantly (P < 0.05) after heating, while
changes were insignicant (P > 0.05) for treatments with GA.
Overall, changes of L* values were less signicant than those of a*
and b* values.
The reduction in redness (positive a* value) of anthocyanin solutions is caused by the loss of avylium cation and the hydrolysis
of a double bond on the C ring of anthocyanin molecules (Goto &
Kondo, 1991). Additionally, the sugar moiety (usually glucose or
sucrose) of anthocyanins can be decomposed during thermal
treatments, and the produced free sugar and glycone (Brouillard &
Delaporte, 1977) accelerate the reduction of redness (Brouillard &
Delaporte, 1977). A recent study reported that water-soluble carbohydrates can signicantly improve the stability of anthocyanins
by the co-pigmentation mechanism, which involves the reduction
of water activity around anthocyanins and therefore the decomposition of sugar from anthocyanins (Sadilova, Stintzing,
Kammerer, & Carle, 2009). This mechanism may also apply to
water-soluble GA that improved the preservation of the color of
anthocyanin solutions after heating (Fig. 1 and Table 1).
3.2. Molecular binding between GA and anthocyanins
When small molecules quench uorescence of biopolymers,
uorescence spectroscopy is a convenient tool to study their interactions. The quenching of uorescence intensity can be dynamic

Fig. 1. Appearance of solutions with 0.51 mg/mL anthocyanins at pH 5.0 with (A, B and C) and without (D, E and F) 10 mg/mL gum arabic before (A and D) and after heating at 80 (B
and E) and 126 (C and F)
C for 30 min.

Y. Guan, Q. Zhong / LWT - Food Science and Technology 64 (2015) 706e712

709

Table 1
Color parameters and concentrations of anthocyanins in 0.51 mg/mL anthocyanin solutions at pH 5.0, before and after heating at 80 and 126
C for 30 min without and with
10 mg/mL gum arabic (GA).a
Parameter

Without GA

56.82
19.54
10.22
0.051

L
aa
ba
Concentration (%w/w)
a

With GA

Heating at 80
C

Before heating
c

0.03
0.02a
0.04f
0.000a

61.03
0.85
8.94
0.047

Heating at 126
C

2.47
0.18e
1.06e
0.004b

66.92
1.96
7.01
0.023

Heating at 80
C

Before heating

0.03
0.01f
0.04a
0.000d

60.02
15.34
5.48
0.051

0.02
0.05b
0.03d
0.000a

60.49
7.91
2.79
0.048

0.57
0.33c
0.51c
0.002b

Heating at 126
C
61.32
7.47
2.99
0.031

0.03b
0.03d
0.02b
0.001c

Numbers are mean standard error from triplicates. Different superscript letters in the same row indicate signicant differences in the mean (P < 0.05).

Table 2
The uorescence quenching constant (kq), hydrodynamic diameter, and zeta potential of gum arabic (GA) with and without anthocyanins at pH 5.0, before and after heating at
80 and 126
C for 30 min.a
kq , 1010 M1s1
Before heating 80
C
b

Anthocyanin-GA 3.99 0.25

GA
N/Db
a
b

Hydrodynamic diameter (nm)

126
C

17.35 0.65
N/D

Before heating
a

16.23 0.08
N/D

80
C

Zeta potential (mV)

126
C

143.95 9.40
140.67 10.28
78.40 14.99b 78.80 7.23b

Before heating
a

80
C

126
C
a

136.90 1.70
22.80 0.44 24.73 1.06 23.95 0.91a
75.95 19.45b 23.92 1.26a 25.08 1.08a 24.50 2.37a

Numbers are mean standard error from triplicates. Different superscript letters in the same parameter indicate signicant differences in the mean (P < 0.05).
N/D: not determined.

or static due to intermolecular collision and complex formation,

respectively, and can be analyzed using the SterneVolmer model
(Eq. (4)) (Borissevitch, 1999). A quenching constant (kq) higher than
2.0  1010 M1s1 indicates the formation of complexes (Lange,
Kothari, Patel, & Patel, 1998).

F0
1 kq t0 Q  1 KSV Q 
F

where, Fo and F are the uorescence intensities without and with

the quencher, [Q] is the concentration of the quencher, Ksv is the

C
0.0

y=-0.0040x
2
R =0.9378

0.0

y=-0.0278x
2
R =0.9324

-0.5

ln(Ct/C0)

-0.1

ln(Ct/C0)

(4)

-0.2

-1.0

-1.5

-0.3
-2.0

-0.4
-2.5

20

40

60

80

20

60

80

Time (min)

Time (min)

D
0.0

0.00

y=-0.0010x
2
R =0.8627

ln(Ct/C0)

-0.04

y=-0.0222x
R2=0.9409

-0.5

-0.02

ln(Ct/C0)

40

-1.0

-1.5

-0.06

-2.0

-0.08
0

20

40

Time (min)

60

80

20

40

60

80

Time (min)

Fig. 2. Degradation kinetics of 0.51 mg/mL anthocyanin solution at pH 5.0 during heating at 80 (A and B) and 126 (C and D)
C without (A and C) and with (B and D) 10 mg/mL gum
arabic.

710

Y. Guan, Q. Zhong / LWT - Food Science and Technology 64 (2015) 706e712

Table 3
The degradation rate constant (k) and half-life of anthocyanins during heating at 80
and 126
C in the 0.51 mg/mL anthocyanin solution with and without 10 mg/mL gum
arabic (GA) at pH 5.0.a
k (102 min1)

Sample

Without GA, 80 C
With GA, 80
C
Without GA, 126
C
With GA, 126
C

0.41
0.16
4.28
2.34

0.15
0.07b
0.82a
0.36b

Half-life (min)
221.47
445.13
16.58
30.14

34.80b
61.26a
2.92b
4.70a

a
Numbers are mean standard error from triplicates. Different superscript letters in the same column and same temperature indicate signicant difference in the
mean (P < 0.05).

SterneVolmer quenching constant, and constant to ( 108 s) is the

life time of uorescence in absence of a quencher (Lange et al.,
1998).
In the present study, the kq determined from the anthocyaninGA dispersions before and after heating at 80 and 126
C for
30 min was 3.99  1010, 17.35  1010, and 16.23  1010 M1s1,
respectively (Table 2), which was signicantly higher than
2.0  1010 M1s1. Therefore, it can be concluded that complexes
formed between anthocyanin and GA molecules at pH 5.0 before
and after thermal treatment. The kq of GA-anthocyanin mixture
increased signicantly (Table 2), indicating the strengthened
binding between the two molecules, after heating.

and 1.8 times at 80 and 126
C. After heating at 80 and 126
C for
30 min, the residual concentration of anthocyanins with GA was
1.02 and 1.35 times higher than that without GA (Table 1). These
results indicated that thermal degradation of anthocyanins was
inhibited by GA after forming complexes.
In a previous study, anthocyanins were co-spray dried with
maltodextrin (MD), GA, or their mixture, and the solutions after
hydrating spray-dried powder and adjusting to pH 3.0 were heated
at 60, 80, and 98
C (Idham, Muhamad, & Sarmidi, 2012). At 60
C,
the degradation rate of anthocyanins in the control solution was
identical to the sample with GA, while improvements were
observed for the MD and MDeGA treatments. At 80
C, the half-life
of anthocyanin degradation in the GA treatment was even shorter
than the control and was shorter than the other two treatments. At
98
C, slight improvements of anthocyanin stability was observed
for all three treatments. As discussed previously, water-soluble
carbohydrates can improve the thermal stability of anthocyanins
by the co-pigmentation mechanism (Sadilova et al., 2009). Additionally, anthocyanins are known to be stable below pH 4.0
(Buchweitz, Speth, Kammerer, & Carle, 2013; Howard, Brownmiller,
Prior, & Mauromoustakos, 2013). Therefore, the co-pigmentation
mechanism may be dominant at pH 3.0, while complex formation
is involved in the present study.

3.4. Changes of particle size and morphology after complex

formation and heating
3.3. Thermal degradation kinetics
The kinetics of anthocyanin degradation at 80 and 126
C with
and without GA is shown in Fig. 2. The degradation of anthocyanins
in solutions with and without GA followed the rst order kinetics,
as reported previously (Krca et al., 2007; Wang & Xu, 2007). The
determined thermal degradation rate constant (k; Eq. (1)) is listed
in Table 3. The k value at 80 and 126
C decreased by 61.0% and
45.3%, respectively, after addition of GA. Correspondingly, the
respective half-life of anthocyanin degradation increased by 2.0

The hydrodynamic diameter of GA increased signicantly

(P < 0.05) after mixing with anthocyanins (Table 2). The hydrodynamic diameter of GA with and without anthocyanins did
not change signicantly after heating at 80 and 126
C for
30 min (Table 2). The structural stability was also observed in
AFM that showed the similar morphology (Fig. 3) and particle
height (Table 4) of GA and anthocyanin-GA samples at pH 5.0
before and after heating. The height of GA tested in the present
study (~2 nm, Table 4) was in agreement with the study that

Fig. 3. AFM topography images of anthocyanin-gum arabic (A, B and C) and gum arabic solutions (D, E and F) at pH 5.0, before (A and D) and after heating at 80 (B and E) and 126 (C
and F)
C for 30 min. Scale bars 1 mm.

Y. Guan, Q. Zhong / LWT - Food Science and Technology 64 (2015) 706e712

80

Table 4
The AFM height of particles in gum arabic solutions with and without anthocyanins
at pH 5.0, before and after heating at 80 and 126
C for 30 min.a

Before heating
Heating at 80
C
Heating at 126
C
Before heating
Heating at 80
C
Heating at 126
C

2.2
2.5
2.4
11.4
11.0
12.2

0.3b
0.1b
0.8b
5.4a
1.8a
4.3a

a
Numbers are mean standard error from two images. Different superscript
letters indicate signicant differences in the mean (P < 0.05).

reported the height of the monomeric GA in aqueous solutions

to be 1e3 nm (Ikeda, Funami, & Zhang, 2005). Additionally, the
height of the anthocyanin-GA particles was about 5e6 times of
the GA (Table 4), suggesting that the particles are composed of
several monomeric GA (Ikeda, Gohtani, Nishinari, & Zhong,
2013; Lv, Yang, Li, Zhang, & Abbas, 2014). The results were in
agreement with our recent study about the stable norbixin-GA
complexes at pH 5.0 after similar thermal treatments (Guan &
Zhong, 2014).
3.5. Zeta-potential of dispersions
The zeta-potential of GA at pH 5.0 with and without anthocyanins before and after heating at 80 and 126
C for 30 min is shown
in Table 2. No signicant differences were observed for the negative
zeta-potential of GA after mixing with anthocyanins (P > 0.05),
suggesting anthocyanins are present in the complex core. Similar
results were observed in our recent study for the complexes of
anthocyanins and mannoprotein (Wu et al., 2015). Because both
mannoproteins and GA are protein-polysaccharide conjugates,
hydrophobic interactions being the complex formation mechanism
concluded previously can also be applied to GA-anthocyanin
complexes. Additionally, heating did not have a signicant impact
on zeta-potentials of both GA and anthocyanin-GA dispersions
(P > 0.05), which suggests the thermal stability of complexes,
agreeing with the previous section based on particle dimension and
morphology.

70
DPPH radical scavenging capacity, %

With anthocyanin

Particle height (nm)

before heating

60

80 C

50

126 C

40
30

20
10
0
anthocyanin

anthocyanin-GA

GA

70
ABTS radical scavenging capacity, %

Without anthocyanin

Treatment

711

60

before heating
80 C

50

126 C
40
30
20
10
0
anthocyanin

anthocyanin-GA

GA

3.6. Antioxidant capacity

The DPPH and ABTS radical scavenging capacities and ferric
reducing power were used to collectively evaluate the antioxidant
capacity of anthocyanins (Cavalcanti et al., 2011). As shown in Fig. 4,
DPPH and ABTS radical scavenging capacities and ferric reducing
power of anthocyanin solutions at pH 5.0 signicantly decreased
after thermal treatment at 80 and 126
C for 30 min. About 89%,
92%, and 92%, and 78%, 63%, and 63% of DPPH and ABTS radical
scavenging capacities and ferric reducing power of anthocyanins
were retained after heating at 80 and 126
C for 30 min, respectively, while these prospective retentions were 95%, 93%, and 97%,
and 92%, 77%, and 95% after adding GA. The increased retentions of
antioxidant properties of anthocyanins by GA were signicant
(P < 0.05), which is likely due to the reduced thermal decomposition of anthocyanins that are included in the complex core. The
observations in Fig. 4 were similar to the preservation of the antioxidant capacity of anthocyanins by yeast mannoprotein after
thermal treatment at pH 7.0 (Wu et al., 2015).
4. Conclusions
GA at 10 mg/mL signicantly improved the stability of 0.51 mg/
mL anthocyanins at pH 5.0 during thermal treatment at 80 and

Fig. 4. DPPH (A) and ABTS (B) radical-scavenging activity (%), and ferric reducing
power (C) of 0.51 mg/mL anthocyanin solutions at pH 5.0 with and without 10 mg/mL
GA, before and after heating at 80 and 126
C for 30 min.

126
C, due to formation of complexes by hydrophobic attraction.

The thermal degradation of anthocyanins with and without GA
followed the rst order kinetics, and the addition of GA effectively
reduced the degradation rate constant by 39.0 and 54.7% and
correspondingly increased the half-life by 2 and 1.8 times after
heating at 80 and 126
C, respectively. Based on light scatting, zeta
potential, and AFM data, multiple GA molecules were involved in

712

Y. Guan, Q. Zhong / LWT - Food Science and Technology 64 (2015) 706e712

the formation of complexes, and the inclusion of anthocyanins in

the complex core resulted in the improved anthocyanin stability.
The reduced thermal degradation of anthocyanins by GA signicantly improved the retention of the antioxidant capacity of anthocyanins after heating. Findings from the present study suggest
the possibility to expand the application of anthocyanins by GA in
thermally-processed foods at around pH 5.0.
Acknowledgments
Authors acknowledge the University of Tennessee and the USDA
National Institute of Food and Agriculture Hatch Project 223984 for
nancial support. We gratefully thank Drs. Yue Zhang, Yangchao
Luo and Gang Liu for the excellent technical support in antioxidant
assays and AFM experiments.
References
Berk, Z. (2008). Food process engineering and technology (1st ed.). London: Academic
Press.
Borissevitch, I. E. (1999). More about the inner lter effect: corrections of
SterneVolmer uorescence quenching constants are necessary at very low
optical absorption of the quencher. Journal of Luminescence, 81(3), 219e224.
Brouillard, R., & Delaporte, B. (1977). Chemistry of anthocyanin pigments. 2. Kinetic
and thermodynamic study of proton transfer, hydration, and tautomeric reactions of malvidin 3-glucoside. Journal of the American Chemical Society, 99(26),
8461e8468.
Buchweitz, M., Speth, M., Kammerer, D. R., & Carle, R. (2013). Impact of pectin type
on the storage stability of black currant (Ribes nigrum L.) anthocyanins in pectic
model solutions. Food Chemistry, 139(1e4), 1168e1178.
Buffo, R., Reineccius, G., & Oehlert, G. (2001). Factors affecting the emulsifying and
rheological properties of gum acacia in beverage emulsions. Food Hydrocolloids,
15(1), 53e66.
Cavalcanti, R. N., Santos, D. T., & Meireles, M. A. A. (2011). Non-thermal stabilization
mechanisms of anthocyanins in model and food systemsdAn overview. Food
Research International, 44(2), 499e509.
Colantuoni, A., Bertuglia, S., Magistretti, M., & Donato, L. (1991). Effects of Vaccinium
myrtillus anthocyanosides on arterial vasomotion. Arzneimittel-Forschung,
41(9), 905e909.
rillon, J.-M. (2009). ComparDudonne, S., Vitrac, X., Coutiere, P., Woillez, M., & Me
ative study of antioxidant properties and total phenolic content of 30 plant
extracts of industrial interest using DPPH, ABTS, FRAP, SOD, and ORAC assays.
Journal of Agricultural and Food Chemistry, 57(5), 1768e1774.
Ersus, S., & Yurdagel, U. (2007). Microencapsulation of anthocyanin pigments of
black carrot (Daucus carota L.) by spray drier. Journal of Food Engineering, 80(3),
805e812.
Gomes, J. F., Rocha, S., Pereira, M. d. C., Peres, I., Moreno, S., Toca-Herrera, J., et al.
(2010). Lipid/particle assemblies based on maltodextrinegum arabic core as
bio-carriers. Colloids and Surfaces B: Biointerfaces, 76(2), 449e455.
Goto, T., & Kondo, T. (1991). Structure and molecular stacking of anthocyaninsdower color variation. Angewandte Chemie International Edition in English, 30(1),
17e33.
Gris, E., Ferreira, E., Falc~
ao, L., & Bordignon-Luiz, M. (2007). Caffeic acid copigmentation of anthocyanins from Cabernet Sauvignon grape extracts in model
systems. Food Chemistry, 100(3), 1289e1296.
Guan, Y., & Zhong, Q. (2014). Gum arabic and Fe2 synergistically improve the heat
and acid stability of norbixin at pH 3.0e5.0. Journal of Agricultural and Food
Chemistry, 62(52), 12668e12677.
Howard, L. R., Brownmiller, C., Prior, R. L., & Mauromoustakos, A. (2013). Improved
stability of chokeberry juice anthocyanins by b-cyclodextrin addition and
refrigeration. Journal of Agricultural and Food Chemistry, 61(3), 693e699.
Idham, Z., Muhamad, I. I., & Sarmidi, M. R. (2012). Degradation kinetics and color
stability of Spray-dried encapsulation anthocyanins from Hibiscus Sabdariffa L.
Journal of Food Process Engineering, 35(4), 522e542.
Ikeda, S., Funami, T., & Zhang, G. (2005). Visualizing surface active hydrocolloids by
atomic force microscopy. Carbohydrate Polymers, 62(2), 192e196.
Ikeda, S., Gohtani, S., Nishinari, K., & Zhong, Q. (2013). High acyl gellan networks
probed by rheology and atomic force microscopy. Food Science and Technology
Research, 19(2), 201e210.
Islam, A., Phillips, G., Sljivo, A., Snowden, M., & Williams, P. (1997). A review of
recent developments on the regulatory, structural and functional aspects of

gum arabic. Food Hydrocolloids, 11(4), 493e505.

Jayme, M., Dunstan, D., & Gee, M. (1999). Zeta potentials of gum arabic stabilised oil
in water emulsions. Food Hydrocolloids, 13(6), 459e465.

lu, B. (2007). Effects of temperature, solid content

Krca, A., Ozkan,
M., & Cemerog
and pH on the stability of black carrot anthocyanins. Food Chemistry, 101(1),
212e218.
Lange, D. C., Kothari, R., Patel, R. C., & Patel, S. C. (1998). Retinol and retinoic acid
bind to a surface cleft in bovine beta-lactoglobulin: a method of binding site
determination using uorescence resonance energy transfer. Biophysical
Chemistry, 74(1), 45e51.
Lertittikul, W., Benjakul, S., & Tanaka, M. (2007). Characteristics and antioxidative
activity of Maillard reaction products from a porcine plasma proteineglucose
model system as inuenced by pH. Food Chemistry, 100(2), 669e677.
Lv, Y., Yang, F., Li, X., Zhang, X., & Abbas, S. (2014). Formation of heat-resistant
nanocapsules of jasmine essential oil via gelatin/gum arabic based complex
coacervation. Food Hydrocolloids, 35, 305e314.
Maier, T., Fromm, M., Schieber, A., Kammerer, D. R., & Carle, R. (2009). Process and
storage stability of anthocyanins and non-anthocyanin phenolics in pectin and
gelatin gels enriched with grape pomace extracts. European Food Research and
Technology, 229(6), 949e960.
Main, J., Clydesdale, F., & Francis, F. (1978). Spray drying anthocyanin concentrates
for use as food colorants. Journal of Food Science, 43(6), 1693e1694.
Marques, L. S., Narita, N. E., Costa, G. G. d., & Rezende, M. C. (2010). Evaluation of
thermal and mechanical behaviors of PPS/carbon ber laminates processed in
autoclave under different consolidation cycles. Polmeros, 20(4), 309e314.
Odriozola-Serrano, I., Soliva-Fortuny, R., & Martn-Belloso, O. (2010). Changes in
bioactive composition of fresh-cut strawberries stored under superatmospheric
oxygen, low-oxygen or passive atmospheres. Journal of Food Composition and
Analysis, 23(1), 37e43.
Oidtmann, J., Schantz, M., M
ader, K., Baum, M., Berg, S., Betz, M., et al. (2012).
Preparation and comparative release characteristics of three anthocyanin
encapsulation systems. Journal of Agricultural and Food Chemistry, 60(3),
844e851.
Pan, K., & Zhong, Q. (2013). Improving clarity and stability of skim milk powder
dispersions by dissociation of casein micelles at pH 11.0 and acidication with
citric acid. Journal of Agricultural and Food Chemistry, 61(38), 9260e9268.
Peterson, J., & Dwyer, J. (1998). Flavonoids: dietary occurrence and biochemical
activity. Nutrition Research, 18(12), 1995e2018.
Re, R., Pellegrini, N., Proteggente, A., Pannala, A., Yang, M., & Rice-Evans, C. (1999).
Antioxidant activity applying an improved ABTS radical cation decolorization
assay. Free Radical Biology and Medicine, 26(9), 1231e1237.
Sadilova, E., Stintzing, F. C., Kammerer, D. R., & Carle, R. (2009). Matrix dependent
impact of sugar and ascorbic acid addition on color and anthocyanin stability of
black carrot, elderberry and strawberry single strength and from concentrate
juices upon thermal treatment. Food Research International, 42(8), 1023e1033.
Santos, D., & Meireles, M. (2009). Jabuticaba as a source of functional pigments.
Pharmacognosy Reviews, 3(5), 127e132.
Valduga, E., Lima, L., Prado, R. d., Padilha, F. F., & Treichel, H. (2008). Extraction,
spray drying and microencapsulating of 'Isabel'grape (Vitis labrusca) bagasse
anthocyanin. Ci^
encia e Agrotecnologia, 32(5), 1568e1574.
Wang, J., & Mazza, G. (2002). Effects of anthocyanins and other phenolic compounds on the production of tumor necrosis factor a in LPS/IFN-g-activated
RAW 264.7 macrophages. Journal of Agricultural and Food Chemistry, 50(15),
4183e4189.
Wang, W.-D., & Xu, S.-Y. (2007). Degradation kinetics of anthocyanins in blackberry
juice and concentrate. Journal of Food Engineering, 82(3), 271e275.
Williams, D. N., Gold, K. A., Holoman, T. R. P., Ehrman, S. H., & Wilson, O. C., Jr.
(2006). Surface modication of magnetic nanoparticles using gum arabic.
Journal of Nanoparticle Research, 8(5), 749e753.
Wu, J., Guan, Y., & Zhong, Q. (2015). Yeast mannoproteins improve thermal stability
of anthocyanins at pH 7.0. Food Chemistry, 172(1), 121e128.
Xiong, S., Melton, L. D., Easteal, A. J., & Siew, D. (2006). Stability and antioxidant
activity of black currant anthocyanins in solution and encapsulated in glucan
gel. Journal of Agricultural and Food Chemistry, 54(17), 6201e6208.
Xu, H., Liu, X., Yan, Q., Yuan, F., & Gao, Y. (2015). A novel copigment of quercetagetin
for stabilization of grape skin anthocyanins. Food Chemistry, 166(1), 50e55.
Zhang, Y., & Zhong, Q. (2012). Binding between bixin and whey protein at pH 7.4
studied by spectroscopy and isothermal titration calorimetry. Journal of Agricultural and Food Chemistry, 60(7), 1880e1886.
Zhang, Y., & Zhong, Q. (2013). Encapsulation of bixin in sodium caseinate to deliver
the colorant in transparent dispersions. Food Hydrocolloids, 33(1), 1e9.
Zhao, C., Giusti, M. M., Malik, M., Moyer, M. P., & Magnuson, B. A. (2004). Effects of
commercial anthocyanin-rich extracts on colonic cancer and nontumorigenic
colonic cell growth. Journal of Agricultural and Food Chemistry, 52(20),
6122e6128.