Food Hydrocolloids 61 (2016) 821e831

Contents lists available at ScienceDirect

Food Hydrocolloids
journal homepage: www.elsevier.com/locate/foodhyd

Development of tannic acid cross-linked hollow zein nanoparticles as

potential oral delivery vehicles for curcumin
Siqi Hu, Taoran Wang, Maria Luz Fernandez, Yangchao Luo*
Department of Nutritional Sciences, University of Connecticut, Storrs, CT, 06269, USA

a r t i c l e i n f o

a b s t r a c t

Article history:
Received 5 May 2016
Received in revised form
21 June 2016
Accepted 6 July 2016
Available online 10 July 2016

Sodium carbonate was proposed as a sacrice template and tannic acid was used as a natural cross-linker
to prepare hollow zein nanoparticles (HZN/T). The formulation of nanoparticles, including the amount of
water, zein and sodium carbonate, were optimized by surface response methodology (Box-Behnken
design). The optimized HZN/T exhibited a small dimension of 87.93 nm with a PDI of only 0.105 and a
zeta potential of
39.70 mV, indicating the nanoparticles were homogenous and formed stable colloidal
dispersion. Then curcumin was used as a model lipophilic nutrient to explore the encapsulation and
delivery potentials of HZN/T, in comparison with hollow zein nanoparticles without tannic acid (HZN/NT)
and solid zein nanoparticles with tannic acid (SZN/T) prepared under the same conditions. Generally, the
encapsulation efciency of HZN/T or HZN/NT was signicantly higher than that of SZN/T. Interestingly,
encapsulation of curcumin dramatically increased particle size of SZN/T by 50 nm, while it did not induce
any expansion of the dimension of HZN/T due to its hollow structure. The molecular interactions between curcumin and zein nanoparticles were investigated by Fourier transform infrared spectroscopy
and uorescent spectrophotometer. The in vitro stability and release prole of nanoparticles were
evaluated under the simulated gastrointestinal conditions. Although all types of zein nanoparticles
showed a sustained release of curcumin, cross-linking via tannic acid played an important role to make
zein nanoparticles more resistant against simulated intestinal digestion. Therefore, compared with
traditional SZN/T, the HZN/T developed in this study has promising features as a potential oral delivery
system for lipophilic nutrients/drugs.
2016 Elsevier Ltd. All rights reserved.

Keywords:
Zein
Hollow nanoparticles
Tannic acid
Cross-link
Curcumin
Encapsulation
Stability

1. Introduction
Curcumin, a member of the polyphenol family, is a hydrophobic
compound found in the herb Curcuma longa. Previous studies
supported that curcumin possesses anti-oxidant (Weber, Hunsaker,
Abcouwer, Deck, & Vander Jagt, 2005), anti-inammatory
(Satoskar, Shah, & Shenoy, 1986), anti-microbial (De et al., 2009),
and anti-carcinogenic properties (Limtrakul, Lipigorngoson,
Namwong, Apisariyakul, & Dunn, 1997). Nevertheless, the
compromised bioavailability, including poor absorption and rapid
clearance from human body, obstructed curcumin to be applied in
therapeutics and functional foods (Anand, Kunnumakkara,
Newman, & Aggarwal, 2007). In order to increase its bioavailability, nanotechnology has been employed to develop delivery

* Corresponding author. Department of Nutritional Sciences, University of Connecticut, 3624 Horsebarn Road Extension, U-4017, Storrs, CT, 06269-4017, USA.
E-mail address: yangchao.luo@uconn.edu (Y. Luo).
http://dx.doi.org/10.1016/j.foodhyd.2016.07.006
0268-005X/ 2016 Elsevier Ltd. All rights reserved.

systems for curcumin. Thus far, a variety of nanoparticles have been

studied as potential vehicles to deliver curcumin through different
routes, either injection (Sun et al., 2013) or oral administration
(Shaikh, Ankola, Beniwal, Singh, & Kumar, 2009). Although oral
administration has been a preferred route over injection, many
delivery systems of curcumin are prepared from either synthetic
polymers and/or surfactants, which are often associated with potential risk of toxicity. Alternatively, food biopolymers, including
polysaccharides and proteins, have received increasing attention
for their encapsulation and delivery potentials for nutrients (Abd
El-Salam & El-Shibiny, 2012; B. Hu & Huang, 2013; McClements,
2015), due to their naturally-occurring status with biodegradability and biocompatibility.
Zein nanoparticles are one of the recently studied biodegradable
polymeric nanoparticles and have been applied in drug delivery
(Lai & Guo, 2011), (Regier, Taylor, Borcyk, Yang, & Pannier, 2012),
nutrient delivery (Luo, Zhang, Whent, Yu, & Wang, 2011) and tissue
engineering (Dong, Sun, & Wang, 2004; Paliwal & Palakurthi, 2014).

822

S. Hu et al. / Food Hydrocolloids 61 (2016) 821e831

Zein is a prolamine protein found as storage protein of maize and is

generally regarded as safe (GRAS) as a food additive (Janes,
Kooshesh, & Johnson, 2002). Compositional analysis indicated
that zein contains two thirds of non-polar amino acids (Righetti,
Gianazza, Viotti, & Soave, ), and one third of polar amino acids
(Geraghty, Peifer, Rubenstein, & Messing, 1981; Luo & Wang, 2014).
Zein nanoparticles are commonly prepared by liquid-liquid antisolvent precipitation method and such-prepared nanoparticles are
often reported to have a dimension of 200e300 nm with a solid
internal core. Recently, Xu and colleagues reported a novel method
to prepare hollow zein nanoparticles (HZN) by introducing sodium
carbonate as a sacrice template (Helan Xu, Jiang, Reddy, & Yang,
2011). The HZN had a particle size around 60 nm and exhibited a
signicantly higher loading capacity than solid zein nanoparticles
(SZN). This is considered as a novel methodology to prepare
protein-based hollow nanoparticles, as the core template sodium
carbonate is removed simultaneously during the formation of zein
nanoparticles by anti-solvent process, while thermal or chemical
treatment is usually required in the preparation of many other
hollow nanoparticles (Peng & Sun, 2007; Zeng et al., 2008).
Nevertheless, for a new methodology, the comprehensive optimization, including fabrication procedures and nanoparticle formulations, is needed to prepare HZN for delivery of nutrients.
As all other protein nanoparticles, a major challenge for zein
nanoparticles as oral delivery vehicles for nutrients is the stability
under gastrointestinal conditions, where high concentration of
salts, extreme pHs, and digestive enzymes are present. Although
zein is relatively resistant to digestive enzymes, it has been shown
that zein nanoparticles without any modication were rapidly hydrolyzed or aggregated (Luo, Teng, & Wang, 2012; Luo et al., 2011),
resulting in a burst effect of encapsulated drugs/nutrients. Many
strategies are available to possibly address this concern. For
instance, surface coating with another polymer (Luo et al., 2012;
Luo et al., 2013; Luo et al., 2011) has demonstrated promising effects to slow release rate of nutrients from zein nanoparticles.
Cross-linking is another approach that may be helpful to stabilize
zein nanoparticles and improve their delivery potentials. Particularly, citric acid has been recently reported as a non-toxic chemical
cross-linker to signicantly prolong in vivo residence time of HZN
(H. Xu, Shen, Xu, & Yang, 2015). However, heating at 50  C for 10 h is
required in this process to create amide bonds between carboxylic
groups of citric acid and amine groups of zein, which may negate its
applications in encapsulating temperature-sensitive labile nutrients, such as curcumin. Previous studies have shown that binding
proteins with tannins increase their stability and thus decreased
digestibility (Taylor, Bean, Ioerger, & Taylor, 2007). Tannic acid belongs to the tannin family and contains abundant hydroxyl groups,
allowing it to form hydrogen bonds with other compounds,
including protein. The formation of tannic acid-protein complex
was suggested to be based on non-covalent interactions between
carbonyl groups of the protein and hydroxyl groups of the tannic
acid (Van Buren & Robinson, 1969). Hydrophobic amino acids in
zein, such as proline and phenylalanine, are potential binding sites
for tannic acid (Jobstl, OConnell, Fairclough, & Williamson, 2004).
Therefore in order to increase the stability of zein nanoparticles in
the gastrointestinal tract, tannic acid could be a possible option to
make zein nanoparticles more stable.
The rst objective of this study was to explore the main and
interactive effects of nanoparticle fabrication conditions and to
optimize the formulation of hollow nanoparticles using surface
response methodology with Box-Behnken design. Tannic acid was
included in the formulation as a non-covalent cross-linker to stabilize HZN/T. The second objective was to evaluate the potential of
as-prepared HZN/T as an oral delivery vehicle for curcumin, with
solid SZN/T as well as HZN without tannic acid (HZN/NT) studied as

controls. The physicochemical properties, including morphology,

intermolecular interactions, encapsulation efciency, as well as
stability and controlled release under simulated gastrointestinal
conditions were comprehensively characterized.
2. Materials and methods
2.1. Materials
Zein, sodium carbonate (Na2CO3, purity  99.0%), and tannic
acid (ACS reagent) were purchased from Sigma-Aldrich Corp. (St.
Louis, MO, USA). Curcumin (purity  98%) was obtained from
ACROS Organics (Geel, Belgium). Other chemicals, including pepsin
and pancreatin, were of analytical grade and obtained from Thermo
Fisher Scientic (Pittsburgh, PA, USA).
2.2. Preparation of nanoparticles
HZN/T was prepared by using anti-solvent precipitation method
with Na2CO3 used as sacricing template (Helan Xu et al., 2011). A
certain amount of zein powder (100e500 mg) was dissolved in
10 mL 70% v/v aqueous ethanol to form a stock solution. Sacrice
template was formed by pouring 0.7 mL of pure ethanol into 0.3 mL
deionized water containing carbonate (2.5e7.5 mg). Then, 1 mL of
zein stock solution was mixed with the template solution. Preestablished amount of deionized water (5e15 mL) with 2 mg tannic acid was pipetted gradually to the mixture under constant
magnetic stirring. Particles were stirred for another 30 min for
stabilization. HZN/NT and SZN/T were prepared in parallel as controls for comparison purpose. In the SZN/T preparation, carbonate
was pre-dissolved in 1 mL water (no formation of sacrice template), rather than 70% v/v aqueous ethanol solution.
2.3. Box-Behnken design
A Box-Behnken design with 15 runs, 3 factors and 3 levels, was
utilized to optimize formulations of HZN/T. Design-Expert software
(Trial Version 8.0.6, Stat-Ease Inc., MN) was used to develop a
quadratic response surface, which was designed for testing main
effects, quadratic effects of independent variables, and interactions
between them. The software generated a quadratic model based on
the following equation:

Y A0 A1 x1 A2 x2 A3 x3 A4 x1 x2 A5 x1 x3 A6 x2 x3
A7 x21 A8 x22 A9 x23
Y represents the response variable needs to be optimized; A1eA9
are regression coefcients of independent factors, showing their
interactions and quadratic terms; X1, X2 and X3 are the coded levels
of explanatory variables. Selection of independent variables, i.e.
amount of additional water (X1), Na2CO3 (X2) and zein (X3), was
based on previous study (Helan Xu et al., 2011), and the detail information was shown Table 1. Independent factors were divided
into three levels:
1, 0 and 1, representing low, medium and high
values, respectively. Table 2 presented the study design matrix
established by the software. Dependent variables of this study

Table 1
Experiment factors and levels.
Factors

Level (
1)

Level (0)

Level (1)

Water (mL), X1
Sodium (mg), X2
Zein (mg), X3

5
2.5
10

10
5
30

15
7.5
50

S. Hu et al. / Food Hydrocolloids 61 (2016) 821e831

823

and observed using SEM with magnication of 20,000 at11.5 kV

(JSM-6330F, JEOL Ltd., Tokyo, Japan).

Table 2
Experiment design (Box-Behnken design).
Run

X1

X2

X3

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15

0
0
1
0
1
1

1

1
0
0
0
1

1

1
0

0
1
0

1
1
0

1
0
0

1
0

1
1
0
1

0

1
1
1
0

1
0
1
0

1
0
0
0

1
1

2.5.3. Lyophilization and redispersion of curcumin-loaded zein

nanoparticles
Freshly prepared curcumin-loaded nanoparticles were rst
frozen at
80  C, and then freeze-dried by Labconco Freezone
(Labconco Corp., Kansas City, MO) under
80  C and vacuum
overnight. The freeze-dried samples were re-dispersed in deionized water at 1 mg/mL, followed by particle size, PDI and zeta potential measurements.

were: particle size (Y1), polydispersity index (PDI) (Y2) and zetapotential (Y3), as described in Table 3.
2.4. Encapsulation of curcumin
Formulations of curcumin loaded zein nanoparticles were based
on the optimal formulation obtained from Box-Behnken design.
The preparation followed the same procedures for abovementioned
zein nanoparticles in Section 2.2, with curcumin being dissolved in
the zein stock solution. Different zein/curcumin mass ratios
(5:1e100:1) were investigated to optimize the encapsulation
efciency.
2.5. Nanoparticle characteristics
2.5.1. Particle size, PDI and zeta potential
Particle size, PDI, and zeta potential of samples were measured
using a dynamic light scattering (DLS) analyzer (Nano Zetasizer ZS,
Malvern Instruments, Ltd., Worcestershire, UK), which was equipped with alternative 50 mW laser beam at a scattering angel of 173
and wavelength of 532 nm or 633 nm. To avoid multiple scattering,
all samples were diluted by 20 times prior to particle size and PDI
measurements. All measurements were performed in three replicates at 25  C.
2.5.2. Scanning electron microscopy (SEM)
The morphology of nanoparticles were observed by scanning
electron microscopy (SEM). Freshly prepared samples were rst
cast-dried on an aluminum pan using a vacuum desiccator (VDC-11,
Jeio Tech, Korea) overnight. Then the samples were mounted onto
specimen stub, coated with a thin layer of gold by a sputter coater

2.5.4. Fourier transform infrared (FT-IR) spectroscopy

FT-IR spectrum analysis was performed to determine the
chemical structures of curcumin-loaded zein nanoparticles and the
individual ingredients (i.e., zein, curcumin and tannic acid). The
freeze-dried samples were then mounted onto ATR crystal for
analysis utilizing a Nicolet iS5 FT-IR spectrometer (Thermo Scientic, Waltham, MA, USA). The spectra were obtained at
500e4000 cm
1 wavenumbers with a resolution of 4 cm
1.

2.5.5. Fluorescence spectroscopy

The binding between curcumin and zein was determined using
uorescence spectrophotometry (Sahu, Kasoju, & Bora, 2008).
Curcumin-loaded zein nanoparticles (loaded with 10% curcumin)
were selected for uorescence measurements using a PerkinElmer
LS 55 uorescence spectrometer (PerkinElmer Instruments, UK).
The excitation wavelength was 420 nm, and the emission spectra
were recorded from 440 to 610 nm. The slit widths were both
10 nm for excitation and emission. Two free curcumin controls
were prepared. One was dissolved and diluted with pure ethanol to
the same curcumin concentration as in nanoparticles, whereas the
other was prepared by pre-dissolving curcumin in ethanol followed
by appropriate dilution with sodium carbonate to achieve equivalent pH and ionic strength, as in the nanoparticles.

2.5.6. Encapsulation efciency

Briey, freshly prepared curcumin-loaded nanoparticles were
centrifuged at 17,217 g for 30 min to remove unstable particles and
excess curcumin. Curcumin in the precipitate was extracted by 1 mL
pure ethanol with vortexing for 1 min. Curcumin concentrations
were determined by a UVeVis spectrophotometer (Evolution 201,
Thermo Scientic, Waltham, MA, USA) at 428 nm with a standard
curve (R2 0.999). The encapsulation efciency (EE) was calculated
according to the following equation:

Table 3
Statistical analysis results of particle size (Y1), PDI (Y2), and zeta potential (Y3).
Parameters*

Intercept
X1
X2
X3
X1*X2
X1*X3
X2*X3
X1*X1
X2*X2
X3*X3
*

Particle size (Y1)

PDI (Y2)

Zeta potential (Y3)

Coefcient

P-value

Coefcient

P-value

Coefcient

P-value

127.03

0.46
16.23
50.69
2.20
4.56

14.47
14.04
13.03
17.67

N/A
0.9141
0.0099
<0.0001
0.7141
0.4580
0.0513
0.0634
0.0784
0.0303

0.081
0.010

0.023
0.025

0.013

0.016
0.003
0.016

0.008
0.036

N/A
0.1420
0.0092
0.0070
0.1594
0.1025
0.7236
0.1141
0.3795
0.0079

32.13

8.41
2.94
2.70
0.17

0.40

0.95
2.93

0.42
1.25

N/A
<0.0001
0.0013
0.0019
0.7948
0.5582
0.1966
0.0069
0.5541
0.1175

X1, amount of water; X2, amount of sodium carbonate; X3, amount of zein.

824

EE%

S. Hu et al. / Food Hydrocolloids 61 (2016) 821e831

Total curcuminmg
Precipitated curciminmg

 100
Total curcuminmg

amount of zein (X3), as indicated in Table 3. The following quadratic

equation depicts the relationships with a R2 0.9757 (Fig. 1A),
indicating a good t:

Y1 127:03
0:46X1 16:23X2 50:69X3 2:2X1 X2

2.5.7. Stability
The stability of curcumin-loaded nanoparticles was evaluated in
simulated gastric and intestinal conditions, according to our previous study (Zhou, Wang, Hu, & Luo, 2016). Briey, 1 mL of freshly
prepared nanoparticles was added to 9 mL of simulated gastric uid
(SGF, pH 2 or pH 4, with 1 mg/mL pepsin) or simulated intestinal
uid (SIF, pH 7, with 10 mg/mL pancreatin), followed by incubation
at 37  C for 2 or 4 h, respectively. Particle size, PDI and zeta potential were measured after each incubation.

4:56X1 X3
14:47X2 X3 14:04X12 13:03X22 17:67X32

As indicated in the above equation and Fig. 2A, both the
amounts of sodium carbonate and zein were positively related to

2.5.8. Controlled release prole

The pH dependent controlled release study was performed by
dialysis membrane method (Shaikh et al., 2009). SGF (pH 4), and SIF
(pH 7) premixed with equal volume of ethanol were prepared as
release medium to create sink conditions for curcumin. Briey,
3.7 mg of freeze-dried curcumin loaded nanoparticle powders
dispersed in 3 mL SGF release medium was loaded into dialysis bag
with a molecular weight cutoff of 12e14 kDa (Fisherbrand, Fisher
Scientic, Pittsburgh. PA). The release was carried out at 37  C
consecutively in SGF for 2 h and then SIF for another 4 h under
moderately shaking. The 1 mL of release medium was extracted
from the reservoir at predetermined time point, and fresh release
medium was replenished. Curcumin concentration was determined
as previously described using UVevis spectrophotometer.
2.6. Statistics
All measurements were performed in triplicate and data were
expressed as mean standard deviation. All data were rst tested
for homogeneity of variances using Levene test, and then One-way
ANOVA with Tukeys Honest Signicant Difference test (if Levene
test negative, P > 0.05) or Games-Howell test (if Levene test positive, P < 0.05) was employed to determine differences among
experimental results. P < 0.05 was considered as signicance level.
The analyses were carried out using SPSS statistical software
package (SPSS, Version 22.0, Chicago, USA).
3. Results and discussion
3.1. Optimization of HZN characteristics using Box-Behnken design
Among all different types of response surface designs, the BoxBehnken design has been found to be more efcient compared to
other designs and has been widely applied in analytical system
optimization (Ferreira et al., 2007). In this study, each experimental
run was measured in triplicate with the amount of tannic acid xed
at 2 mg, and the average value was recorded. After collecting all
experimental data, the design-expert software suggested that
quadratic model was the best t model. The software generated
three-dimensional graphs to depict relationships between amount
of ingredients and each characteristic of HZN/T. In the 3-D plot, one
factor was xed at a constant value when main effects and interactions of the other two factors were being analyzed. Comparing
to statistical analysis with only numbers, these graphs exhibited
changes in the response surface in a more vivid form.
3.1.1. Particle size (Y1)
Particle size of HZN/T ranged from 88 to 219 nm, signicantly
dependent upon the amount of sodium carbonate (X2) and the

Fig. 1. Correlations between predicted and actual results of surface response method.

S. Hu et al. / Food Hydrocolloids 61 (2016) 821e831

825

Fig. 2. Selected surface response plots showing (A) Effect of sodium amount and zein amount on particle size, with water amount xed at level 0, 10 mL; (B) Effect of Na2CO3
amount and zein amount on PDI, with water amount xed at level 0, 10 mL; (C) Effect of Na2CO3 amount and zein amount on zeta potential, with water amount xed at level 0,
10 mL; (D) Effect of water amount and Na2CO3 amount on zeta potential, with zein amount xed at level 0, 30 mg.

particle size, with the amount of zein exhibiting a much greater

impact, however, water did not have a signicant effect. Generally,
when the zein concentration was low, the particle size remained
stable with a xed zein concentration, due to the relatively small
size of carbonate crystals being formed during precipitation process. Nevertheless, particle size of HZN/T rapidly increased thereafter as the amount of sodium carbonate further increased,
especially when the zein concentration was low. This may be
explained by the excessive amount of sodium carbonate in the
dispersion causing the aggressive growth of crystal size due to
supersaturation. A similar observation was also reported in the case
of hollow silica nanoparticles, showing that the shape and size of
hollow nanoparticles depended on the sizes of the internal sacrice
templates (Chen, Ding, Wang, & Shao, 2004). In contrast, when zein
concentration was high, such effect of templates was weakened.
This was probably owing to the insufcient templates for zein
nanoparticles and thus the formation of larger solid particles
became possible. On the other hand, the particle size of HZN/T
increased in a much more rapid behavior with the increase of zein
concentration, regardless of the amount of templates. It is notable
that increasing zein concentration in the dispersion induced formation of thicker coating around the templates, resulting in the
rapid growth of particle size of HZN/T.

3.1.2. PDI (Y2)

PDI of HZN/T (Y2) varied from 0.054 to 0.177, and the correlation
between PDI and independent factors can be well explained by this
equation with a R2 0.9326 (Fig. 1B):

Y2 0:081 0:009875X1
0:023X2 0:025X3
0:013X1 X2

0:016X1 X3 0:003X2 X3 0:016X12
0:008042X22
0:036X32
As shown in above equation, the coefcient of X1 was extremely
low, and thus it is clear that the amount of water was not an
important variable for determining the PDI. On the other hand,
both X2 and X3 were signicant affecting factors because their coefcients were relatively high. Generally, the PDI decreased with
increasing of sodium carbonate but increased with increasing of
zein concentration, as reected by the negative and positive value
of their respective coefcient. A higher PDI indicated a less homogeneous distribution of particle size when zein concentration
was high but the template amount was low. Nevertheless, it is
worth mentioning that all the PDI values were smaller than 0.3,
indicating that all samples were considered as homogeneous
nanoparticles with mono-distributed dimensions.

826

S. Hu et al. / Food Hydrocolloids 61 (2016) 821e831

Table 4
Predicted and actual results of surface response method.
Sets

Size (nm)

PDI

Zeta potential (mV)

Predicted
Actual

77.48
87.93 1.73

0.120
0.105 0.023

39.94

39.70 1.51

Table 5
Particle size, PDI, and surface charge of different samples based on the optimized
formulation.
Sample*
HZN/T
HZN/NT
SZN/T

Size (nm)

PDI
a

87.93 1.73
92.44 2.24a
230.40 1.85b

Zeta potential (mV)

a

0.105 0.023
0.09 0.010a
0.173 0.054a

39.70 1.51

36.93 1.55a

31.03 1.33b

*
HZN/T, hollow zein nanoparticles with tannic acid; HZN/NT, hollow zein nanoparticles without tannic acid; SZN/T, solid zein nanoparticles with tannic acid. Data
in the same column with different superscript letter were signicantly different
(p < 0.05).

3.1.3. Zeta potential (Y3)

As illustrated in Fig. 2C and D, zeta potential of HZN/T ranged
from
16.9 to
41.5 mV and the statistical analysis indicated that it
was signicantly affected by all three factors (Table 3). The relationships between zeta potential and independent factors are
dictated by the following equation with a R2 0.9891 (Fig. 1C)
showing a good t:

Y2
32:13
8:14X1 2:94X2 2:70X3 0:17X1 X2

0:40X1 X3
0:95X2 X3 2:93X12
0:42X22 1:25X32
Compared with the amount of zein and sodium carbonate, the
amount of water exhibited much more dramatic effects on the zeta
potential. The effect of sodium carbonate was primarily explained
by the ionic screening effect on the electrical double layer of zein
molecules (Hunter, 2013). Although serving as a sacrice template
for the formation of hollow structure, the addition of sodium carbonate also introduced a lot of salts as sodium ions were released
from solubilized crystals upon pouring the mixture into water
during the liquid-liquid dispersion process. The increase of ionic
strength caused screening effect, thus reducing the surface potential of proteins. This may also dictate the drastic effect of water
amount on zeta potential. Increasing the volume of water during
liquid-liquid dispersion process, both concentrations of zein and
sodium carbonate were greatly reduced, and therefore the extent of
screening effect was consequently attenuated, resulting in a signicant increase of magnitude of zeta potential.

3.1.4. Validation of optimized formulation

Optimization of HZN/T properties was conducted by the Designexpert software with goals of achieving minimum particle size,
minimum PDI and highest absolute value of zeta potential. The
optimized formulation provided by the software was 11.45 mL of
additional water, 11.2 mg of zein and 2.5 mg of sodium carbonate.

To verify the validity of the optimized formulation, three batches of

HZN/T were fabricated based on the optimized formulation, and the
particle size, PDI and zeta potential were measured and compared
with theoretical values. As shown in Table 4, the actual particle size,
PDI and zeta potential were 87.93 1.73 nm, 0.105 0.023 and

39.70 1.51 mV, respectively, all of which were in a good
agreement with predicted results.
3.2. Effect of tannic acid on zein nanoparticles
To evaluate the function of tannic acid cross-linking on zein
nanoparticles, the basic particulate characteristics were compared
among HZN/T, HZN without tannic acid (HZN/NT), as well as SZN
with tannic acid (SZN/T) as a control (Table 5). The HZN/T and HZN/
NT had much smaller particle size compared with the SZN/T.
Meanwhile, it is worth mentioning that although not statistically
signicant, the PDI values of both HZN/T and HZN/NT were lower
than that of SZN as well. Similar observation was reported in a
previous study, which attributed the smaller size of HZN to the
effect of pre-formed sodium carbonate crystals acting as heterogeneous nucleation sites, thus preventing the formation of large
particles during dispersion process (Helan Xu et al., 2011). On the
other side, tannic acid did not exhibit apparent effect on particle
size and PDI. The SZN/T had a size of 230.40 nm with a negative zeta
potential of
31.03 mV, which were both in a good agreement with
literatures (Luo et al., 2013; Luo et al., 2011; Zhong & Jin, 2009). The
zeta potential of HZN/NT and HZN/T was
36.93 and
39.70 mV,
respectively, being signicantly greater than that of SZN. This may
be in part explained by the smaller size of hollow structured particles, thus providing a larger surface area which increased electric
charges per surface area unit (Yin Win & Feng, 2005).
3.3. Encapsulation and delivery potentials of HZN for curcumin
3.3.1. Characterization of curcumin-loaded zein nanoparticles
To optimize the encapsulation of curcumin in HZN/T, ve
loading percentages were studied and the results are shown in
Table 6. Compared with blank HZN/T particles, the particle size
remained unchanged (87e99 nm) until curcumin loading reached
20%, when it signicantly increased to 114 nm. This indicated that
entrapment of curcumin caused a signicant expansion of the
hollow core when the loading was too high, which might also
explain the saturation of curcumin in the core at 20% loading. This
hypothesis was further conrmed by the EE results that a signicant reduction in EE at 20% loading. The high zeta potential
together with small PDI values evidenced that the stable nanoparticles with narrow size distribution were obtained. Considering
all characteristics as a whole, the HZN/T with 10% curcumin loading
was selected as the optimal percentage for further experiments.
Consequently, curcumin was also encapsulated into the HZN/NT
and SZN/T at 10% loading to compare the effects of tannic acid and
hollow/solid structure on the particle characteristics. As shown in
Table 7, while the addition of tannic acid had a negligible impact,
the SZN/T with a solid core exhibited a signicantly greater particle

Table 6
Particle size, PDI, surface charge, and encapsulation efciency of samples loaded with different amount of curcumin.
Curcumin loading (%)

Size (nm)

PDI

Zeta potential (mV)

EE (%)

1
2
4
10
20

86.69 3.66a
87.37 4.14a
90.27 4.27a
99.55 5.28a
114.20 7.76b

0.122 0.065a
0.097 0.017a
0.091 0.010a
0.101 0.027a
0.088 0.021a

45.0 3.64a

40.1 0.83ab

38.9 1.85ab

35.0 0.90b

38.1 2.78b

98.22 0.07a
98.41 0.14a
98.32 0.26a
95.82 0.56b
88.26 0.29c

*Data in the same column with different superscript letter were signicantly different (p < 0.05).

S. Hu et al. / Food Hydrocolloids 61 (2016) 821e831

Table 7
Particle size, PDI, surface charge, and encapsulation efciency of different samples
loaded with equivalent amount of curcumin.
Sample* Size (nm)
HZN/T
HZN/NT
SZN/T

PDI

827

formulation (Fig. 4B and D), which may be owing to the conformational change in protein structure induced by the interactions
between zein and tannic acid (Zou et al., 2015).

Zeta potential (mV) EE (%)

0.101 0.027a
35.0 0.90a

99.55 5.28a
94.66 4.63a
0.097 0.010a
37.10 1.74a
282.47 29.92b 0.141 0.069a
34.30 1.51a

95.82 0.56a
97.23 0.60b
93.49 0.22c

*
HZN/T, hollow zein nanoparticles with tannic acid; HZN/NT, hollow zein nanoparticles without tannic acid; SZN/T, solid zein nanoparticles with tannic acid. Data
in the same column with different superscript letter were signicantly different
(p < 0.05).

size and a lower EE, suggesting the advantage of hollow structure

on the encapsulation of nutrients over solid structure.
The morphology of nanoparticles was observed under SEM as
shown in Fig. 3. Generally, the hollow structured nanoparticles
(HZN/T and HZN/NT, Fig. 3A and B, respectively) were smaller than
the one with solid core (SZN/T, Fig. 3C). However, compared with
the dimensions determined by DLS (Table 7), it is also apparent that
nanoparticles with tannic acid, i.e. HZN/T and SZN/T, had a significantly larger size. This might be in part owing to that tannic acid, as
a cross-linker, induced aggregation and growth of particles,
because its local concentration increased dramatically as water
evaporated during cast-drying process before SEM observation. Our
hypothesis is in line with a previous study reporting that the signicant aggregation of zein nanoparticles occurred with a high
concentration of tannic acid in the formulation (Zou, Guo, Yin,
Wang, & Yang, 2015). In addition, the hollow nanoparticles were
very homogenous and well separated (Fig. 3A and B), while the
solid nanoparticles tended to clump together without clear partition (Fig. 3C). Similar observations have also been reported in
previous studies that zein nanoparticles prepared with low-energy
input techniques (without high speed homogenization or microuidization) were more prone to aggregate (Luo et al., 2012; Zhong
and Jin, 2009; Zou et al., 2015).

3.3.2. Interactions between curcumin and zein

FT-IR was used to investigate the intermolecular interactions
among different ingredients in the nanoparticles, and the spectra
are shown in Fig. 4. Apparently, no signicant shift was found in the
vibration peaks ranging from 3200 to 3400 cm
1, indicating no
dramatic change of hydrogen bonding in zein (Luo et al., 2012).
However, in the spectra of the HZN/T and SZN/T samples (Fig. 4B,
D), the vibration peaks at 758/759 cm
1 could be assigned to C]C
in benzene rings of tannic acid (Socrates, 2004), indicating its
presence in the particles. The peaks at 832/833 and 880 cm
1 were
found in all nanoparticles, attributing to out-of-plane CCH group of
aromatic ring connected with enolic and keto part of curcumin
molecule (K. Hu et al., 2015; Kolev, Velcheva, Stamboliyska, &
Spiteller, 2005) and implying the presence of curcumin. Interestingly, peaks at 1113 cm
1 and 1272 cm
1 in the curcumin spectra
(data not shown), representing aromatic rings of curcumin (K. Hu
et al., 2015), were shifted to 1121e1123 cm
1 and
1281e1284 cm
1, respectively. This indicated the hydrophobic interactions between zein and curcumin. With the presence of tannic
acid, new peaks at 1199 cm
1 and 1317 cm
1 were observed (Fig. 4B
and D), which probably resulted from hydrophobic interactions
between pentagalloyl glucose of tannic acid and proline residues of
zein (Baxter, Lilley, Haslam, & Williamson, 1997). Zou et al. (2015)
also reported that incorporation of tannic acid in the formulation
of zein-based Pickering emulsion gels induced the formation of
noncovalent bonds that strengthened gel structure. Additionally,
another noticeable shift was the amide I bond at 1645 cm
1 of zein
molecules (Fig. 4A) when tannic acid was added in the nanoparticle

Fig. 3. SEM images of cast-dried curcumin-loaded zein nanoparticles: (A) Hollow zein
nanoparticles with tannic acid (HZN/T); (B) Hollow zein nanoparticles without tannic
acid (HZN/NT); (C) Solid zein nanoparticles with tannic acid (SZN/T).

828

S. Hu et al. / Food Hydrocolloids 61 (2016) 821e831

Fig. 4. Fourier transform infrared spectroscopy (FT-IR) spectra of different samples. (A) Zein powder; (B) Hollow zein nanoparticles with tannic acid (HZN/T); (C) Hollow zein
nanoparticles without tannic acid (HZN/NT); (D) Solid zein nanoparticles with tannic acid (SZN/T).

The molecular binding between curcumin and zein in different

samples was further investigated by uorescence spectroscopy, as
presented in Fig. 5. The uorescence of curcumin that was
dispersed in pure ethanol had a peak at 529 nm, whereas curcumin
that was pre-dissolved in ethanol and then diluted with water to
the equivalent nal ethanol concentration and pH value in nanoparticles (about 10% and 9.5, respectively) did not show any uorescence intensity. As previously reported, the photochemical
properties and uorescence spectra of curcumin signicantly
depend on the environmental polarity and pH (Bong, 2000;
Chignell et al., 1994). Therefore this discrepancy was caused by
dramatic decrease in the environmental hydrophobicity and increase in the pH value. Nevertheless, compared with the curcumin
controls, curcumin in all three nanoparticles exhibited signicant
uorescence intensities with the peaks being blue shifted to
522 nm, indicating a more hydrophobic and neutral environment
provided by the encapsulation. These ndings are in agreement
with the literature (Pan, Zhong, & Baek, 2013; Sahu et al., 2008).
Moreover, the uorescence intensity of the SZN/T was slightly
higher than that of the HZN/T and HZN/NT. This may be explained
by the fact that the solid internal core provides more hydrophobic
microenvironment with additional binding sites for curcumin and
thus gave rise to the uorescence intensity.
3.3.3. Re-dispersibility of freeze-dried powders
Lyophilization has been considered as an effective method to
improve long-term storage stability of nanoparticles, however,
many nanoparticles made from hydrophobic polymers (such as
zein) are no longer able to re-constitute in water after freezedrying. Zein nanoparticles prepared by traditional liquid-liquid
dispersion method cannot be re-constituted in water, unless

aqueous/ethanol solution is used or high concentration surfactant

is present. In this study, all three types of curcumin-encapsulated
zein nanoparticles were freeze-dried and then re-dispersed in
deionized water at a concentration of 1 mg/mL to evaluate physiochemical properties of re-constituted products. Compared to
original characteristics (Table 7), the particle size was reduced by

Fig. 5. Fluorescence emission spectra of free curcumin dispersed in pure ethanol

(control 1), free curcumin diluted with sodium carbonate solution (control 2), hollow
zein nanoparticles with tannic acid (HZN/T), hollow zein nanoparticles without tannic
acid (HZN/NT), and solid zein nanoparticles with tannic acid (SZN/T).

S. Hu et al. / Food Hydrocolloids 61 (2016) 821e831

140 and 10 nm for solid and hollow structured nanoparticles,

respectively (Table 8). The PDI value in re-constituted SZN/T was
signicantly greater than the both hollow nanoparticles. Particle
size and PDI of the reconstituted nanoparticles have been recognized as signicant parameters for examining quality loss during
freeze-drying process (Abdelwahed, Degobert, Stainmesse, & Fessi,
2006). Therefore, the reduced particle size with elevated PDI
Table 8
Characteristics of redispersed lyophilized powder loaded with equivalent amount of
curcumin (1 mg/mL).
Sample*

Size (nm)

PDI

Zeta potential (mV)

HZN/T
HZN/NT
SZN/T

89.99 4.75a
86.06 2.06a
143.0 4.65b

0.111 0.014a
0.094 0.017a
0.248 0.027b

40.10 1.71a

37.90 2.65a

39.80 1.10a

*
HZN/T, hollow zein nanoparticles with tannic acid; HZN/NT, hollow zein nanoparticles without tannic acid; SZN/T, solid zein nanoparticles with tannic acid. Data
in the same column with different superscript letter were signicantly different
(p < 0.05).

Table 9
Particle size, PDI, and surface charge of samples after digestion under pH 2, 4 and 7.
pH

Sample*

Size (nm)

PDI

Zeta potential (mV)

HZN/T
HZN/NT
SZN/T
HZN/T
HZN/NT
SZN/T
HZN/T
HZN/NT
SZN/T

1791.25 372.08a
1275.5 137.94a
1346.25 250.40a
99.78 2.48a
90.54 6.65a
300.67 2.42b
879.90 60.34a
419.90 24.29b
1099.33 30.89c

0.325 0.058a
0.613 0.033b
0.386 0.110a
0.092 0.009a
0.146 0.086ab
0.245 0.005b
0.337 0.040a
0.563 0.055b
0.331 0.019a

6.05 0.49a
0.22 0.81b
4.86 0.07c

25.50 7.26a

25.00 7.87a

32.23 7.45a

13.67 0.503a

13.57 0.569a

13.10 0.436a

*HZN/T, hollow zein nanoparticles with tannic acid; HZN/NT, hollow zein nanoparticles without tannic acid; SZN/T, solid zein nanoparticles with tannic acid.
Different superscript letter denoted signicant difference among the data in the
same column under the same pH condition (p < 0.05).

829

suggested the quality loss and less stability of solid nanoparticles

after freeze-drying process. Moreover, the smaller count rates of
reconstituted solid nanoparticles further conrmed the particle
loss or poor re-dispersibility after freeze-drying (data not shown).
3.3.4. Stability in simulated gastrointestinal conditions
To explore the potential as oral delivery vehicles, the stability of
three nanoparticles was determined in simulated gastric and intestinal conditions, respectively. Three conditions were tested, i.e.
fast-status (pH 2) and fed-status (pH 4) gastric conditions with
pepsin as digestive enzyme, and intestinal condition (pH 7) with
pancreatin as digestive enzyme. As shown in Table 9, incubation at
fast-status gastric condition for 2 h induced severe aggregation in
all three samples, ending up with precipitates and microparticles
larger than 1 mm with PDI > 0.3. As revealed in Fig. 6, it is proposed
that the destabilization of zein nanoparticles was mainly due to the
protein aggregation/denaturation at extremely acidic pH, rather
than the enzymatic digestion by pepsin, because the disintegration
of zein nanoparticles by pepsin was not observed. This can be
deduced from the size distribution curve that only one peak of
larger size was observed (Fig. 6B). Nevertheless, when incubated at
fed-status gastric conditions, all nanoparticles exhibited better
stability. Although the particle sizes of all samples were similar to
the original particles, the size distribution of hollow nanoparticles
exhibited a better prole than solid ones (Fig. 6C). The increased
ratio of small particles in SZN/T indicated that the partial digestion
took place causing the disintegration of larger particles to smaller
ones (Teng, Li, & Wang, 2014). On the other hand, under intestinal
condition, the nanoparticles with tannic acid (HZN/T and SZN/T)
showed a better resistant to digestion by pancreatin, which was
able to degrade a-zein dimers while pepsin was not (Hurtadopez & Murdan, 2006). The HZN/NT particles without tannic
Lo
acid were disintegrated into multiple smaller fractions (Fig. 6D),
explaining its smaller mean particle size and larger PDI value
(Table 9).

Fig. 6. The DLS curves of (A) Original curcumin-loaded zein nanoparticles; (B) curcumin-loaded zein nanoparticles after digestion under pH 2 for 2 h; (C) curcumin-loaded zein
nanoparticles after digestion under pH 4 for 2 h; (D) curcumin-loaded zein nanoparticles after digestion under pH 7 for 4 h. HZN/T, hollow zein nanoparticles with tannic acid; HZN/
NT, hollow zein nanoparticles without tannic acid; SZN/T, solid zein nanoparticles with tannic acid.

830

S. Hu et al. / Food Hydrocolloids 61 (2016) 821e831

3.3.5. Kinetic release of curcumin

Based on the stability data, zein nanoparticles were not stable at
pH 2, therefore the kinetic release of curcumin was only studied at
pH 4 and 7 (Fig. 7). The diffusion of free curcumin across the dialysis
membrane was a convex upward shaped curve, with 70% detected
at pH 4 followed another 15% at pH 7. Curcumin in all three
nanoparticles shared similar release prole that only about 30% of
the total curcumin was released at pH 4 and another 10e15% was
released at pH 7. Burst release was not observed, indicating they all
had good controlled release properties. Interestingly, no dramatic
difference was found between the hollow and the solid samples.
We hypothesized that although hollow nanoparticles were more
stable than solid ones, their smaller size and greater surface area as
well as thinner zein layer may facilitate the release of curcumin
from the core. Similar results were also reported in a previous study
on the release prole of metformin from HZN and SZN (Helan Xu
et al., 2011).

4. Conclusions
In this study, HZN/T with diameters less than 100 nm were
developed by exploring sodium carbonate as a sacricing template
and the formulation was comprehensively optimized by surface
response methodology. Particle size, PDI and zeta potential of HZN/
T were signicantly inuenced by the amount of sodium carbonate
and zein, while the amount of water only had a signicant effect on
zeta potential. Compared with SZN/T, HZN/T had smaller particle
size, higher encapsulation efciency and better stability upon
reconstitution. Encapsulation in zein nanoparticles provided curcumin a strong hydrophobic microenvironment in aqueous solution and hydrophobic interactions were considered as the major
driving force for the encapsulation. In terms of the stability under
simulated GI conditions, tannic acid cross-linked zein nanoparticles
were more resistant against digestion by pancreatin than noncrosslinked ones. All zein nanoparticles demonstrated controlled
release properties for curcumin in simulated GI conditions, but no
dramatic difference was found between solid and hollow nanoparticles. Thus, tannic acid cross-linked zein hollow nanoparticles
would be a suitable vehicle for encapsulation and oral delivery of
lipophilic nutrients/drugs.

Fig. 7. The kinetic release proles of free curcumin and curcumin-loaded zein nanoparticles with different composition under simulated gastric and intestinal conditions.
HZN/T, hollow zein nanoparticles with tannic acid; HZN/NT, hollow zein nanoparticles
without tannic acid; SZN/T, solid zein nanoparticles with tannic acid.

Acknowledgement
This work was supported by the Startup funds from University
of Connecticut, Storrs. The SEM studies were performed using the
facilities in the UConn/FEI Center for Advanced Microscopy and
Materials Analysis (CAMMA).
References
Abdelwahed, W., Degobert, G., Stainmesse, S., & Fessi, H. (2006). Freeze-drying of
nanoparticles: Formulation, process and storage considerations. Advanced Drug
Delivery Review, 58(15), 1688e1713.
Abd El-Salam, M. H., & El-Shibiny, S. (2012). Formation and potential uses of milk
proteins as nano delivery vehicles for nutraceuticals: A review. International
Journal of Dairy Technology, 65(1), 13e21.
Anand, P., Kunnumakkara, A. B., Newman, R. A., & Aggarwal, B. B. (2007).
Bioavailability of curcumin: Problems and promises. Molecular Pharmaceutics,
4(6), 807e818.
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(18), 5566e5577.
Bong, P.-H. (2000). Spectral and photophysical behaviors of curcumin and curcuminoids. Bulletin-Korean Chemical Society, 21(1), 81e86.
Chen, J. F., Ding, H. M., Wang, J. X., & Shao, L. (2004). Preparation and characterization of porous hollow silica nanoparticles for drug delivery application.
Biomaterials, 25(4), 723e727.
Chignell, C. F., Bilskj, P., Reszka, K. J., Motten, A. G., Sik, R. H., & Dahl, T. A. (1994).
Spectral and photochemical properties of curcumin. Photochemistry and
Photobiology, 59(3), 295e302.
De, R., Kundu, P., Swarnakar, S., Ramamurthy, T., Chowdhury, A., Nair, G. B., et al.
(2009). Antimicrobial activity of curcumin against Helicobacter pylori isolates
from India and during infections in mice. Antimicrobial Agents and Chemotherapy, 53(4), 1592e1597.
Dong, J., Sun, Q., & Wang, J.-Y. (2004). Basic study of corn protein, zein, as a
biomaterial in tissue engineering, surface morphology and biocompatibility.
Biomaterials, 25(19), 4691e4697.
Ferreira, S. L., Bruns, R. E., Ferreira, H. S., Matos, G. D., David, J. M., Brandao, G. C.,
et al. (2007). Box-behnken design: An alternative for the optimization of
analytical methods. Analytica Chimica Acta, 597(2), 179e186.
Geraghty, D., Peifer, M. A., Rubenstein, I., & Messing, J. (1981). The primary structure
of a plant storage protein: Zein. Nucleic Acids Research, 9(19), 5163e5174.
Hu, B., & Huang, Q.-R. (2013). Biopolymer based nano-delivery systems for
enhancing bioavailability of nutraceuticals. Chinese Journal of Polymer Science,
31(9), 1190e1203.
Hu, K., Huang, X., Gao, Y., Huang, X., Xiao, H., & McClements, D. J. (2015). Core-shell
biopolymer nanoparticle delivery systems: Synthesis and characterization of
curcumin fortied zein-pectin nanoparticles. Food Chemistry, 182, 275e281.
Hunter, R. J. (2013). Zeta potential in colloid science: Principles and applications (Vol.
2). Academic press.
pez, P., & Murdan, S. (2006). Zein microspheres as drug/antigen carriers:
Hurtado-Lo
A study of their degradation and erosion, in the presence and absence of enzymes. Journal of Microencapsulation, 23(3), 303e314.
Janes, M. E., Kooshesh, S., & Johnson, M. G. (2002). Control of listeria monocytogenes
on the surface of refrigerated, ready-to-eat chicken coated with edible zein lm
coatings containing nisin and/or calcium propionate. Journal of Food Science,
67(7), 2754e2757.
Jobstl, E., OConnell, J., Fairclough, J. P., & Williamson, M. P. (2004). Molecular model
for astringency produced by polyphenol/protein interactions. Biomacromolecules, 5(3), 942e949.
Kolev, T. M., Velcheva, E. A., Stamboliyska, B. A., & Spiteller, M. (2005). DFT and
experimental studies of the structure and vibrational spectra of curcumin. International Journal of Quantum Chemistry, 102(6), 1069e1079.
Lai, L. F., & Guo, H. X. (2011). Preparation of new 5-uorouracil-loaded zein nanoparticles for liver targeting. International Journal of Pharmacy, 404(1e2),
317e323.
Limtrakul, P., Lipigorngoson, S., Namwong, O., Apisariyakul, A., & Dunn, F. W. (1997).
Inhibitory effect of dietary curcumin on skin carcinogenesis in mice. Cancer
Letter, 116(2), 197e203.
Luo, Y., Teng, Z., & Wang, Q. (2012). Development of zein nanoparticles coated with
carboxymethyl chitosan for encapsulation and controlled release of vitamin D3.
Journal of Agricultural and Food Chemistry, 60(3), 836e843.
Luo, Y., & Wang, Q. (2014). Zein-based micro- and nano-particles for drug and
nutrient delivery: A review. Journal of Applied Polymer Science, 131(16), 40696.
Luo, Y., Wang, T. T. Y., Teng, Z., Chen, P., Sun, J., & Wang, Q. (2013). Encapsulation of
indole-3-carbinol and 3,30 -diindolylmethane in zein/carboxymethyl chitosan
nanoparticles with controlled release property and improved stability. Food
Chemistry, 139(1e4), 224e230.
Luo, Y., Zhang, B., Whent, M., Yu, L. L., & Wang, Q. (2011). Preparation and characterization of zein/chitosan complex for encapsulation of alpha-tocopherol, and
its in vitro controlled release study. Colloids and Surfaces B: Biointerfaces, 85(2),
145e152.
McClements, D. J. (2015). Nanoscale nutrient delivery systems for food applications:

S. Hu et al. / Food Hydrocolloids 61 (2016) 821e831

Improving bioactive dispersibility, stability, and bioavailability. Journal of Food
Science, 80(7), N1602eN1611.
Paliwal, R., & Palakurthi, S. (2014). Zein in controlled drug delivery and tissue engineering. Journal of Controlled Release, 189, 108e122.
Pan, K., Zhong, Q., & Baek, S. J. (2013). Enhanced dispersibility and bioactivity of
curcumin by encapsulation in casein nanocapsules. Journal of Agricultural and
Food Chemistry, 61(25), 6036e6043.
Peng, S., & Sun, S. (2007). Synthesis and characterization of monodisperse hollow
Fe3O4 nanoparticles. Angewandte Chemie, 119(22), 4233e4236.
Regier, M. C., Taylor, J. D., Borcyk, T., Yang, Y., & Pannier, A. K. (2012). Fabrication and
characterization of DNA-loaded zein nanospheres. Journal of Nanobiotechnology,
10, 44.
Righetti, P. G., Gianazza, E., Viotti, A., & Soave, C.. Heterogeneity of storage proteins
in maize. Planta, 136(2), 115e123.
Sahu, A., Kasoju, N., & Bora, U. (2008). Fluorescence study of the curcumin
casein
micelle complexation and its application as a drug nanocarrier to cancer cells.
Biomacromolecules, 9(10), 2905e2912.
Satoskar, R. R., Shah, S. J., & Shenoy, S. G. (1986). Evaluation of anti-inammatory
property of curcumin (diferuloyl methane) in patients with postoperative
inammation. International Journal of Clinical Pharmacolology, Therapy, and
Toxicology, 24(12), 651e654.
Shaikh, J., Ankola, D. D., Beniwal, V., Singh, D., & Kumar, M. N. V. R. (2009). Nanoparticle encapsulation improves oral bioavailability of curcumin by at least 9fold when compared to curcumin administered with piperine as absorption
enhancer. European Journal of Pharmaceutical Sciences, 37(3e4), 223e230.
Socrates, G. (2004). Infrared and Raman characteristic group frequencies: Tables and
charts. John Wiley & Sons.
Sun, J., Bi, C., Chan, H. M., Sun, S., Zhang, Q., & Zheng, Y. (2013). Curcumin-loaded
solid lipid nanoparticles have prolonged in vitro antitumour activity, cellular
uptake and improved in vivo bioavailability. Colloids and Surfaces B: Biointerfaces, 111, 367e375.
Taylor, J., Bean, S. R., Ioerger, B. P., & Taylor, J. R. (2007). Preferential binding of

831

sorghum tannins with g-karin and the inuence of tannin binding on karin
digestibility and biodegradation. Journal of Cereal Science, 46(1), 22e31.
Teng, Z., Li, Y., & Wang, Q. (2014). Insight into curcumin-loaded b-lactoglobulin
nanoparticles: Incorporation, particle disintegration, and releasing proles.
Journao of Agricultral and Food Chemistry, 62(35), 8837e8847.
Van Buren, J. P., & Robinson, W. B. (1969). Formation of complexes between protein
and tannic acid. Journal of Agricultral and Food Chem, 17(4), 772e777.
Weber, W. M., Hunsaker, L. A., Abcouwer, S. F., Deck, L. M., & Vander Jagt, D. L.
(2005). Anti-oxidant activities of curcumin and related enones. Bioorganic and
Medincal Chemistry, 13(11), 3811e3820.
Xu, H., Jiang, Q., Reddy, N., & Yang, Y. (2011). Hollow nanoparticles from zein for
potential medical applications. Journal of Materials Chemistry, 21(45),
18227e18235.
Xu, H., Shen, L., Xu, L., & Yang, Y. (2015). Controlled delivery of hollow corn protein
nanoparticles via non-toxic crosslinking: In vivo and drug loading study. Biomed Microdevices, 17(1), 8.
Yin Win, K., & Feng, S.-S. (2005). Effects of particle size and surface coating on
cellular uptake of polymeric nanoparticles for oral delivery of anticancer drugs.
Biomaterials, 26(15), 2713e2722.
Zeng, H., Cai, W., Liu, P., Xu, X., Zhou, H., Klingshirn, C., et al. (2008). ZnO-based
hollow nanoparticles by selective etching: Elimination and reconstruction of
metal
semiconductor interface, improvement of blue emission and photocatalysis. ACS Nano, 2(8), 1661e1670.
Zhong, Q., & Jin, M. (2009). Zein nanoparticles produced by liquideliquid dispersion. Food Hydrocolloids, 23(8), 2380e2387.
Zhou, M., Wang, T., Hu, Q., & Luo, Y. (2016). Low density lipoprotein/pectin complex
nanogels as potential oral delivery vehicles for curcumin. Food Hydrocolloids, 57,
20e29.
Zou, Y., Guo, J., Yin, S. W., Wang, J. M., & Yang, X. Q. (2015). Pickering emulsion gels
prepared by hydrogen-bonded zein/tannic acid complex colloidal particles.
Journal of Agricultral and Food Chemistry, 63(33), 7405e7414.