- Email: [email protected]
The effect of soy protein structural modification on emulsion properties and oxidative stability of fish oil microcapsules
Accepted Manuscript Title: The effect of soy protein structural modification on emulsion properties and oxidative stability of fish oil microcapsules Author: Yating Zhang Chen Tan Shabbar Abbas Karangwa Eric Xiaoming Zhang Shuqin Xia Chengsheng Jia PII: DOI: Reference:
S0927-7765(14)00232-X http://dx.doi.org/doi:10.1016/j.colsurfb.2014.05.006 COLSUB 6411
To appear in:
Colloids and Surfaces B: Biointerfaces
Received date: Revised date: Accepted date:
24-2-2014 28-4-2014 2-5-2014
Please cite this article as: Y. Zhang, C. Tan, S. Abbas, K. Eric, X. Zhang, S. Xia, C. Jia, The effect of soy protein structural modification on emulsion properties and oxidative stability of fish oil microcapsules, Colloids and Surfaces B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014.05.006 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1
The effect of soy protein structural modification on emulsion properties and
2
oxidative stability of fish oil microcapsules Yating Zhang, Chen Tan, Shabbar Abbas, Karangwa Eric, Xiaoming Zhang *, Shuqin Xia, Chengsheng Jia
4
State Key Laboratory of Food Science and Technology, School of Food Science and Technology, Jiangnan
5
University, Lihu Road 1800, Wuxi, Jiangsu 214122, China.
us
cr
ip t
3
* Corresponding author: Tel: +86 510 85197217; Fax: +86 510 85884496.
7
E-mail address: [email protected] (X. Zhang)
Ac ce p
te
d
M
an
6
1
Page 1 of 33
8
9
Abstract
Hydrolysates of soy protein isolate-maltodextrin (SPI-Md) conjugate were used as wall material to
11
prepare fish oil microcapsules by freeze-drying method. Effects of the protein structural modifications on the
12
physicochemical properties of the emulsion and the oxidative stability of the microcapsules were
13
characterized. Compared with emulsions of SPI-Md conjugates or soy protein isolate/maltodextrin (SPI/Md)
14
mixture, lower droplet size (212.5-329.3 nm) and polydispersity index (PDI) (0.091-0.193) were obtained in
15
the fish oil emulsions prepared by SPI-Md conjugate hydrolysates. The improved amphiphilic property of
16
SPI-Md conjugate hydrolysates was supported by the results of surface and interfacial tension, and further
17
confirmed by the improved emulsion stability during the storage period. Although the microencapsulation
18
efficiency (MEE) of SPI-Md conjugate hydrolysates slightly decreased from 97.84% to 91.47% with the
19
increasing degree of hydrolysis (DH), their oxidative stabilities (Peroxide value and headspace propanal)
20
were apparently improved compared with native SPI/Md mixture or SPI-Md conjugates system. Moreover,
21
favourable thermal stability as well as a porous and uniform surface structure of the microcapsules coated by
22
SPI-Md conjugate hydrolysates (DH 2.9%) was observed via the thermal analysis and scanning electron
23
microscope (SEM) micrographs, respectively.
24
Key words
26
cr
us
an
M
d
te
Ac ce p
25
ip t
10
Microcapsules; Soy protein isolate; Conjugates; Limited hydrolysis
1. Introduction
27
Fish oil is known to be a rich source of long chain polyunsaturated fatty acids such as the omega-3 (n-3)
28
family, and many health benefits have been associated with their regular consumption. However, n-3 fatty 2
Page 2 of 33
29
acids are susceptible to oxidative deterioration, which restricts their use in foods [1]. Especially,
30
hydroperoxides, the primary product of lipid oxidation have been considered to be toxic [2].
31
Microencapsulation was reported to be a desirable technique for preventing the oxidation of n-3 fatty acids
32
[1, 3-5]. The selection of wall materials, which significantly affected the stability of powders, was mainly based on
34
their interfacial functionality [6]. The water solubility and amphiphilic properties, the ability to self-associate
35
and interact with variety of substances, and the high molecular chain flexibility of modified proteins were all
36
proved conducive to emulsification [6]. Vegetable proteins are relatively cheap source of wall material for
37
active component microencapsulation, besides their nutritional value and biodegradability. As there is
38
pressing need for multi-functional materials, different modification techniques have been developed to
39
enhance and diversify the protein functionalities, in order to make them be more suitable wall materials for
40
the current microencapsulation techniques.
M
an
us
cr
ip t
33
It was reported that the glycosylated proteins had good emulsifying [7] as well as antioxidant properties
42
[8]. Augustin research group [9-12] confirmed that glycosylated proteins could be used as effective
43
encapsulating materials in the microencapsulation of oils. Akhtar and Dickinson [13] demonstrated that the
44
whey protein–maltodextrin conjugates possessed excellent emulsifying properties and could be used as an
45
alternative to gum arabic. Recently, Drusch et al. [14] reported that the Maillard reaction in aqueous solution
46
could lead to an increase in redox-active compounds of the caseinate–glucose syrup, thus increasing the
47
oxidative stability of encapsulated fish oil. Moreover, Ryszard Amarowicz [15] summarized that enzymatic
48
hydrolysis of proteins was another powerful tool in the modification of their functional properties in food
49
systems. Limited hydrolyzed proteins were reported to exhibit improved emulsifying, solubility and higher
50
antioxidant activities than that of the non-hydrolyzed protein, due to the stretched protein chains [16-18].
51
Nevertheless, single modification method may not be enough to fulfill the pressing demand for
Ac ce p
te
d
41
3
Page 3 of 33
multi-functional materials. Recently, a novel method of protein modification through the limited enzymatic
53
hydrolysis of protein-polysaccharide conjugates has attracted considerable interest. For instance, the sodium
54
caseinate-maltodextrin conjugate hydrolysates were reported to have potential as a water soluble low
55
molecular weight emulsifier and an emulsion stabilizer [19]. Similar results were reported by Zhang et al.
56
[20], which confirmed that the combination of protein-polysaccharide conjugation and controlled enzymatic
57
hydrolysis was a useful strategy to prepare an effective emulsifier.
cr
ip t
52
However, few studies mentioned the application of these hydrolyzed protein-polysaccharide conjugates as
59
wall materials for microencapsulation, especially for fish oil. This study was conducted to evaluate the effect
60
of SPI modifications on the emulsion properties and the oxidative stability of microencapsulated fish oil.
61
Limited enzymatic hydrolysis was carried out following the preparation of soy protein isolate-maltodextrin
62
conjugates through Maillard reaction. Freeze drying was selected mainly due to its lower drying temperature,
63
which could help to protect the fish oil from the oxidative deterioration. It was expected to get further
64
insights into the structural modifications of wall material and its influence on the microencapsulation
65
properties, thus, offering more reference information for proteins modification in the future.
66
2. Materials and Methods
67
2.1 Materials
Ac ce p
te
d
M
an
us
58
68
The defatted soybean meal was obtained from Anyang Mantianxue Food Manufacturing Company
69
(Henan, China). Maltodextrin (Md) with DE values of 8-10, was obtained from Baolingbao Biology
70
Company (Shangdong, China). The enzyme Neutrase was obtained from Novozymes (Jiangsu, China). The
71
fish oil contained approximately 70% EPA/DHA (20/50). All other chemicals used in this study were of
72
analytical grade.
4
Page 4 of 33
73
2.2 Preparation of the Soy Protein Isolate (SPI) SPI was prepared according to the method of Petruccelli and Anon [21]. The defatted soybean meal
75
powder was suspended in 15-fold water and adjusted to pH 7 with 2M NaOH. After stirring for 1 h, the
76
suspension was centrifuged at 8000 g for 30 min and the supernatant was subjected to isoelectric
77
precipitation by adjusting pH to 4.5 with 2M HCl. The protein precipitate, which was recovered by
78
centrifuging (8000 g, 30 min), was re-suspended in water and adjusted to pH 7 with 2M NaOH. After
79
removing small amount of insoluble substances by centrifuging at 8000 g for 30 min, the protein solution
80
was freeze dried and ground to yield SPI powder. All procedures were carried out at room
81
temperature. Crude protein content of SPI determined by Kjeldhal method was 96.7% (w/w) (N×6. 25).
82
2.3 Hydrolysis after glycosylation
M
an
us
cr
ip t
74
SPI (4%, w/v, protein) was incubated with maltodextrin at the SPI/maltodextrin ratio of 2/1 (w/w) for 80
84
min at 80 oC, pH 7.0, and the degree of glycosylation (DG) was 33% determined according to the modified
85
o-phthaldialdehyde (OPA) method [22].
86
The temperature of the glycosylated solution was raised to 54 oC, pH 7.0, followed by enzymatic hydrolysis
87
with Neutrase (Enzyme/Substance (E/S) = 0.5%, 2%, 4%, 6%, 8%) for 25 min to DH 1.8%, DH 2.9%, DH
88
4.6%, DH 5.7%, DH 6.2%, respectively, through pH-stat method [23]. The progress of the reaction process
89
was monitored by the consumption of 0.5 M NaOH. Afterwards, the enzyme was inactivated by heating at
90
80 oC for 20 min. These samples were lyophilized and referred as SPI-Md conjugate hydrolysates.
91
2.4 Emulsion preparation
Ac ce p
te
d
83
92
The modified products of SPI were initially dispersed in warm (70 °C) distilled water. The pH of the
93
resulting solution was adjusted to 7.0 with 1 M NaOH. The oil phase was heated to 70 °C in a water bath
94
prior to coarse emulsion preparation through homogenizer (ULTRA-TURRAX T-25, IKA, Staufen, 5
Page 5 of 33
95
Germany) at 10,000 rpm for 2 min. The coarse emulsion was subjected to two-stage homogenization (35+10
96
MPa) using a high pressure homogenizer (NS1001 L2K, A S Co. Ltd., Parma, Italy).
97
2.5 Emulsion size The particle size distribution of the emulsion droplets was measured by using a Malvern Mastersizer 2000
99
(Malvern Instruments Ltd., Malvern, England). The emulsion was diluted with distilled water. A relative
100
refractive index ηoil/ηwater = 1.095 (ηoil = 1.465 and ηwater = 1.330) was used for the calculation of particle
101
size distribution, assuming that all droplets were spherical in shape. A polydisperse model was used to
102
analyze the data while the intensity of scattered light was taken as an indication of particle size.
103
2.6 Surface and interfacial tension
an
us
cr
ip t
98
The influence of different concentrations of the modified SPI products on the interfacial tension at the fish
105
oil-water phase and the air- water surface tension were measured by the Wilhelmy plate method with an
106
automated tensiometer DCAT21 (DataPhysics instrument GmbH, Germany) at 25 oC. The sample
107
concentration varied between 0.5% and 4.0% (w/w) at pH 7.0.
108
2.7 Rheological measurements of emulsions
Ac ce p
te
d
M
104
109
Rheological measurements of fish oil emulsions were performed at 25 °C according to the method of
110
Liang et al. [24] using the Advanced Rheometric Expansion System (ARES, TA Instruments, New Castle,
111
DE, USA) with a cone and plate geometry (cone diameter = 50 mm, angle = 4°, gap = 0.05 mm). A thin
112
layer of silicone oil was applied on the surface of the samples in order to prevent evaporation. For each
113
measurement, 1.5 mL of the emulsion sample was loaded on the rheometer. The viscosity of nanoemulsions
114
was measured by a steady state flow program with the shear rate ranging from 0 s−1 till steady. Experimental
115
flow curves were fitted to a power law model 6
Page 6 of 33
116
η = Kγn-1
(1)
where η was the viscosity (Pa·s), γ was the shear rate (s−1), K was the consistency index (Pa·sn), and n was
118
the index that provided information about the flow behavior related to the effect of shear rate. There exist
119
three value ranges for n: n < 1 for a shear-thinning fluid, n = 1 for a Newtonian fluid, and n > 1 for a
120
shear-thickening fluid.
121
2.8 Preparation of the Microcapsule powders
us
cr
ip t
117
Emulsions were frozen with liquid nitrogen and subjected to a freeze-drying process (Labconco FreeZone
123
4.5L, freeze-drier, Kansas City, MO, USA). The operating condition of freeze-dryer was −50 °C for a period
124
of up to 24 h for all samples. After the drying process, a fine dry powder was obtained.
125
2.9 Extraction of free oil
d
M
an
122
Extraction of the free oil was made with petroleum ether (b.p. 60-90 oC) following the method as
127
described by Anna Millqvist-Fureby (2003) [25], which was modified to suit the sample volume. One gram
128
of powder was added to 10 ml dried petroleum ether, and shaken for 2 min. Solvent was then separated by
129
filtration. The solid residue was washed with 2×2 ml petroleum ether. The combined filtrate was evaporated
130
using a rotary evaporator. Fat residue was then dried at 105 oC until a constant weight was reached.
131
2.10 Extraction of total oil
Ac ce p
te
126
132
The extraction of total oil was based on the method of Utai Klinkesorn et al. with some modification [26].
133
Five-milliliters of distilled water (60 oC) was added to 0.5 g powder and shaken for 15 min using thermostat
134
Water Bath Vibrator (CD3192, Xutemp, Hangzhou, China). The resulting solution was then extracted with
135
25 mL hexane/isopropanol (3:1, v/v). The tubes were then vortexed for 15 min, and centrifuged for another 7
Page 7 of 33
15 min at 8000 g. The clear organic phase was collected while the aqueous phase was re-extracted with the
137
solvent mixture [3, 4]. After filtration through anhydrous Na2SO4, solvent was evaporated in a rotary
138
evaporator (RE-52AA, Shanghai Biochemical Instrument Company, China) at 70 oC. The solvent-free
139
extract was dried at 105 oC. The amount of encapsulated oil was determined gravimetrically.
140
2.11 Microencapsulation efficiency (MEE)
The encapsulation efficiency (EE) was calculated as follows:
142
EE (%) = (Total oil - free oil)/Total oil×100
an
(2)
2.12 Powder storage
M
143
us
141
cr
ip t
136
Five grams of powder was placed in 20 mL loosely capped amber glass bottle. Samples were stored at 35
145
°C for 4 and 8 weeks in the dark. The extent of lipid oxidation in the powder was monitored by measuring
146
the Peroxide value (POV) and headspace propanal.
147
2.13 Peroxide value (POV)
Ac ce p
te
d
144
148
Hydroperoxides, the primary oxidation products, were measured according to AOAC Official Method
149
965.33 [27], with small modification to suit small sample volumes. 15 mL CH3COOH–CHCl3 was added
150
into 0.50 ± 0.05 g test portion, and the mixture was stirred well to destroy the encapsulants. Then 0.05 mL
151
saturated KI solution was added from Thermal pipet, followed by occasional shaking 1 min, and 30 mL H2O
152
was added. It was slowly titrated with 0.01M Na2S2O3 with vigorous shaking until yellow disappear.
153
Subsequently, 0.05 mL 1% starch solution was added, titration and vigorously shaking was continued to
154
release all I2 fromCHCl3 layer, until blue just disappeared.
155
Peroxide value (milliequivalent peroxide/kg oil or fat) = S×M×1000/g sample, where S = mL, Na2S2O3 8
Page 8 of 33
156
(blank corrected); M = molarity, Na2S2O3 solution.
157
2.14 Headspace propanal A PerkinElmer Model Autosystem XL capillary gas chromatograph (GC) fitted with a DB1 fusedsilica
159
capillary column (25 m × 0.32 mm, 5 µm film thickness; Agilent Technologies, Forest Hill, Victoria,
160
Australia) and an flame
161
the powder sample was weighed and sealed in a 20 mL headspace vial before being equilibrated at 60 °C for
162
15 min. Approximately 2.5 mL of the headspace vapor was injected into the column. The column
163
temperature was increased initially from 60 °C to 75 °C at the rate of 3 °C/min, then to 90 °C at the rate of 5
164
°C/min and finally to 230 °C at the rate of 25 °C/min, where it was held for 20 min. The detector
165
temperature was 240 °C. Headspace propanal content was reported as absolute GC area.
166
2.15. Thermal analysis
ip t
158
d
M
an
us
cr
ionization detector (FID) was used for propanal headspace analysis. One gram of
The thermogravimetric curves (TG) were obtained with the thermal analysis system Q500IR (TA
168
Instruments, USA). Samples were heated from 30 to 900 oC using open aluminium crucibles with
169
approximately 6.0 mg of the sample under a synthetic air flow of 150 mL min-1 at a heating rate of 10 oC
170
min-1. The instrument was preliminarily calibrated with standard mass and with standard calcium oxalate
171
monohydrate. The derivative thermogravimetric curves (DTG), the first derivative of TG curves, were
172
calculated.
173
2.16. Scanning Electron Microscopy (SEM) Measurement.
Ac ce p
te
167
174
Scanning electron micrographs of freeze-dried samples were recorded with a Quanta-200 scanning
175
electron microscope (FEI Co., Eindhoven, The Netherlands) at an accelerating voltage of 10 kV. The
176
powders were sprinkled onto double-backed cellophane tape attached to an aluminum stub before coating 9
Page 9 of 33
177
with gold-palladium in an argon atmosphere.
178
2.17. Statistical analysis The experiments were performed in triplicate and values were expressed as mean ± standard error (SD).
180
All data were subjected to analysis of variance (ANOVA), and the differences between means were
181
evaluated by Duncan’s multiple range test.
182
3. Results and Discussion
183
3.1 Characterization of emulsions prior to drying.
184
3.1.1 Effect of fish oil/protein ratio on the preparation of emulsion
M
an
us
cr
ip t
179
For oil/water (O/W) emulsions, it is always desirable to contain high core volume in order to increase the
186
concentration of hydrophobic active ingredients. Whereas, fish oil content can also affect the emulsion
187
droplet sizes through a coalescence phenomenon [24]. Thus, the fish oil/protein is an important parameter to
188
evaluate the stability of O/W emulsions produced.
Ac ce p
te
d
185
189
Emulsions with various fish oil/protein ratios stabilized by combined modified SPI products (DH 5.7%)
190
were prepared. Formulations and processing conditions for the emulsion preparation are shown in Table 1.
191
The increase of fish oil/protein ratio resulted in a progressive increase of the total solids content, which was
192
the main factor responsible for viscosity increase. However, the change of viscosity had no effect on the
193
preparation of the emulsions as the viscosity was still low. The Z-average diameter and size distributions of
194
emulsions prepared by different fish oil/protein ratios are presented in Table 1. The particle size was slightly
195
changed when the fish oil/protein ratio was in the range of 0.25-1. However, when the fish oil/protein ratio 10
Page 10 of 33
reached to 1.5, a clear increase of the particle size from 282.1 nm to 407.4 nm was observed due to the
197
coalescence [28]. In this case, the excessive oil volume fraction could not be covered by the emulsifier, and
198
part of the free oil droplets would affect the droplet diameter of the emulsion [24].
ip t
196
The polydispersity index (PDI) value represents the particle size distribution of the droplets. A small PDI
200
value indicates a narrow particle size distribution [24]. According to the results shown in Table 1, the PDI
201
values for emulsions stabilized by the modified SPI of DH 5.7% were normally < 0.4, indicating that all of
202
the emulsions had a relatively narrow range of size distribution. The increasing trend of the PDI was
203
apparent, as the oil/protein ratio increased from 0.67 to 1.0. Emulsions showed a small particle size (235.5
204
nm) and a narrow size distribution (PDI, 0.162) at oil to the protein mass ratio of 0.67.
205
3.1.2 Influence of combined modifications on the emulsion properties
M
an
us
cr
199
One of the most important factors leading to efficient microencapsulation of active compounds is their
207
effective dispersion in the wall material solution [28]. The retention of active material during drying process
208
was reported to be increased with the decrease of mean emulsion oil droplet size [8].
Ac ce p
te
d
206
209
The Z-average diameter and PDI of emulsions stabilized by different modified SPI products with the
210
oil/protein ratio of 0.67 are shown in Table 2. It can be seen that the Z-average diameter and PDI of
211
emulsions showed a significant (P < 0.05) increase with the increase of DH. This can be explained by the
212
fact that although limited enzymatic hydrolysis could help to expose the hydrophobic groups and enhance
213
the amphiphilic characteristics, further hydrolysis resulted in shorter peptides, which could affect on the
214
formation of the bulkier polymeric layer around the oil droplets and the stabilization of the oil emulsion [29].
215
Overall, emulsion stabilized by the hydrolyzed conjugates of DH 2.9% displayed desirable emulsifying
216
properties. This observation was consistent with the results reported by Chen et al., which stated that 11
Page 11 of 33
217
excessive hydrolysis may destroy the steric structure of the amphiphilic molecule, which contributed to the
218
emulsifying properties [30]. However, the Z-average diameter values are smaller than those from previous studies on native vegetable
220
proteins used as wall material for microencapsulation [6, 8]. The more efficient dispersion behavior of fish
221
oil (resulting in smaller droplets) can be explained by the presence of stretch proteinic chains after
222
hydrolysis or by improved protein active surface properties after glycosylation, which resulted in the
223
enhancement of the molecular flexibility [6]. Small emulsion Z-average diameter resists against coalescence
224
during the drying process and make emulsions more stable [31]. Therefore, the combined modification
225
method seemed to be efficient for improving the stability and properties of emulsion.
an
us
cr
ip t
219
Additionally, to assess the extended stability of these emulsions, 3 weeks storage test was carried out at 4
226 227
o
228
and the Z-average diameter and PDI kept consistent, indicating an improved storage stability of the
229
combined modified SPI oil-in-water emulsions (Table 2).
230
3.1.3 The surface and interfacial tension
Ac ce p
te
d
M
C. Over this period of storage, there was no obvious phase separation or creaming observed for all samples
231
Changes in the molecular structure of soy protein as a consequence of combined modification may lead to
232
specific surface characteristics or interfacial properties. Fig. 1 shows the surface tension and interfacial
233
tension of native SPI as well as the combined modified SPI products with various degree of hydrolysis (DH),
234
respectively. It was observed that both the surface and interfacial tension decreased with increasing
235
concentration, thus, reflecting better surface and interfacial properties. A slight decrease in surface tension
236
was observed (Fig. 1A) after limited hydrolysis (DH 1.8%, 2.9% and 4.6%) of the glycosylated SPI products.
237
However, the combined modified samples at higher degree of hydrolysis (DH 5.7%-6.2%) showed similar
238
surface tension with the non-hydrolyzed products. This behavior was in agreement with the results of Karina
239
et al. (2009), who pointed out that a low degree of hydrolysis (2–5%) would be enough to improve the 12
Page 12 of 33
240
surface activity of soy protein [32]. High degree of hydrolysis probably decreased the interaction ability of
241
the modified soy protein products at the air-water interface and caused a decrease in foaming ability. The
242
direct relationship between the foam formation, stability and the interfacial properties of adsorbed protein
243
films was reported by Marta research group [33, 34]. The interfacial behavior is an important aspect of both emulsion formation and stability. Proteins are
245
surface-active substances used to stabilize the emulsions by adsorbing at oil-water interface, thereby,
246
reducing the interfacial tension, and forming the mechanical energy barrier at the interface against
247
coalescence in the emulsion system [35]. As shown in Fig. 1B, there was no remarkable difference in the
248
interfacial tension of the native SPI, the glycosylated SPI products and the SPI-Md conjugates hydrolysates
249
(DH 1.8%). However, the limited hydrolysis of DH 2.9% led to a slight decrease of the interfacial tension
250
from 4.058±0.026 mN/m (the glycosylated SPI products ) to 3.354±0.026 mN/m (DH 2.9%) at the
251
equilibrium state, which may be due to the unfolding of the molecule and the increase of protein
252
hydrophobicity which provides a better adsorption process at the oil-water interface [30]. However, further
253
hydrolysis (DH 4.6%-6.2%) resulted in an obvious increase of the interfacial tension compared with the
254
non-enzymatic products. This phenomenon was corresponding to the discussion above on the emulsion
255
properties.
256
3.1.4 The rheological properties of the emulsions
Ac ce p
te
d
M
an
us
cr
ip t
244
257
The rheological properties of the emulsions stabilized by modified SPI products were also investigated to
258
evaluate their emulsion stability. Most emulsion systems used in foods and beverages show shear-thinning
259
behavior, which is important for lowering the viscosity under flow during consumption [24]. The shear rate
260
dependence of apparent viscosities of fish oil emulsions was investigated as shown in Fig. 2. Table 3 shows 13
Page 13 of 33
the rheological parameters fitted with a power law function for emulsions prepared with modified SPI
262
products. The square of the correlation index (R2) was >0.95, suggesting that the model is suitable for the
263
emulsions studied in this work. The flow behavior index (n) indicates the extent of the shear-thinning
264
behavior [36]. Similar typical shear-thinning behaviors were observed for all emulsions (Fig. 2)
265
corresponding to the decreasing tendency of the flow behavior indices (n), which were <1.0 (Table 3).
266
However, there was a slight increase of K values for the emulsions with the increasing DH of modified SPI
267
products, in agreement with the increase of viscosity (Table 2). The change of rheological properties was
268
mainly due to the change in droplet mean diameter [24]. These observations were consistent with the
269
previous report, suggesting that the emulsions of smaller sizes had much higher viscosities than the coarse
270
emulsions and the shear-thinning effect was much stronger in the case of smaller-sized emulsions [37]. As a
271
whole, these characteristics were typical for non-aggregated emulsion droplets, thus, indicating the
272
emulsions stability with no visual instability such as creaming or separation during the storage.
273
3.2 Characterization of the Microcapsule powders
Ac ce p
te
d
M
an
us
cr
ip t
261
274
The microencapsulation efficiency, oxidative stability (headspace propanal and POV), thermal stability
275
and morphology of freeze-dried powders prepared from combined modified SPI products were investigated
276
(Table 4).
277
3.2.1 Microencapsulation efficiency (MEE).
278
The microencapsulation efficiency (MEE) of the powders encapsulated by SPI-Md conjugate hydrolysates
279
was much higher than that of the powders encapsulated by the mixture of SPI and Md, or their conjugates
280
(Table 4). However, a slight decreasing tendency of MEE from 97.84% to 91.47% was observed with 14
Page 14 of 33
increasing DH from 1.8% to 6.2%. That was possibly due to the fact that the shorter chain length of wall
282
material cannot produce a sufficient strong structural matrix to encapsulate the fish oil. Proteinic chains
283
accumulate at the air/water interface of emulsion droplets during drying and thus represent the surface of
284
formed powder particles [38]. Furthermore, the oil droplet size is associated with the microencapsulation
285
efficiency, and small droplet size is expected to achieve high microencapsulation efficiency [14].
286
Additionally, increased oil droplet size in the emulsion might result in higher amount of surface oil.
287
3.2.2 Oxidative stability.
us
cr
ip t
281
To assess the oxidative stability of fish oil powders, an extended storage test (4 weeks or 8 weeks, 35 oC)
289
was carried out. The formation of the hydroperoxides and propanal were measured during the storage time
290
(Table 4).
M
an
288
The fish oil alone was highly oxidizable as the rate of lipid oxidation was very fast. After two-week
292
storage in the same condition, the contents of the POV and headspace propanal soared from 0.65 mmol/kg
293
and 127.5 initially to 157349 mmol/kg and 18757, respectively (Data not shown). However, although the
294
yield of the oxidation products increased with increasing DH, the oxidative stabilities of fish oil powders
295
encapsulated by the SPI-Md conjugate hydrolysate were generally superior to the corresponding powders
296
prepared with SPI-Md conjugates or the SPI/Md mixture, as observed by the lower peroxide value (POV)
297
and headspace propanal (Table 4). Moreover, the formation rate of those lipid oxidation products apparently
298
decreased in the microcapsules coated by SPI-Md conjugate hydrolysates during the storage. The
299
hydroperoxide and propanal content after four-week storage amounted to 4.72 mmol/kg oil and 2688 in the
300
microcapsules prepared by SPI-Md conjugate hydrolysates (DH 2.9%), while the corresponding products in
301
the sample encapsulated with the SPI-Md conjugates amounted to 24.06 mmol/kg oil and 10496.89,
302
respectively. After eight-weeks storage, the oxidative stability of fish oil powders coated by the SPI-Md
303
conjugate hydrolysate were found also better than those prepared with SPI-Md conjugates or the SPI/Md
Ac ce p
te
d
291
15
Page 15 of 33
304
mixture. Therefore, the combined modified wall material after limited hydrolysis had favorable protective
305
action for the core material. Factors such as the interfacial film thickness and robustness, the emulsion parameters and their ability to
307
scavenge oxygen and free radicals, were all reported to play roles in oxidative stability [8]. However, in our
308
study, the degrees of hydrolysis (DH) may not be enough to produce sufficient redox compounds for the
309
antioxidant action. We supposed that the main antioxidant effect was due to the oxygen barrier effect of the
310
combined modified SPI products by forming the strong surface films surrounding the oil droplet, as well as
311
the tightly packed and arranged hydrogen-bonded network structure of the polysaccharide component,
312
which was consistent with the conclusion of Atarés et al [39, 40]. Especially, the improved emulsifying
313
property of the SPI-Md conjugate hydrolysates after limited hydrolysis was conducive to the
314
microencapsulation through decreasing the oil droplet size, preventing the coalescence during the drying
315
process and making microcapsules more complete. Many authors confirmed that retention of active
316
compounds during the drying process could be enhanced by reducing the emulsion Z-average diameter of
317
dispersed components during emulsification [6, 41, 42]. Our work has demonstrated that combined modified
318
SPI products can be used as effective emulsifying and encapsulating materials for the protection of the
319
oxidizable fish oil.
320
3.2.3 Thermal analysis
Ac ce p
te
d
M
an
us
cr
ip t
306
321
Thermogravimetric analysis (TG) and Derivativ Thermogravimetry (DTG) graphs for microparticles
322
encapsulated by the SPI-Md conjugate hydrolysate (DH 2.9%), SPI-Md conjugate, the mixture of SPI and
323
Md, as well as the fish oil are shown in Fig. 3. No degradation of fish oil was observed below 150 oC.
324
However, almost 100% mass loss for fish oil with complete decomposition occurred in the range between
325
200 and 250 °C, and with maximal weight loss at 210 °C as shown in Fig. 3D-F. From the TG and DTG
326
curves of the fish oil microparticles stabilized by the SPI-Md conjugate hydrolysate (DH 2.9%)，slight mass 16
Page 16 of 33
loss related to residual water molecules was observed below 200 oC. The visible decrease of the TG curves
328
occurred at 250 to 500 oC, which was attributed to the degradation of modified SPI products and some core
329
materials. The degradation of the wall material has been found with whole mass loss from 200 to 500 oC
330
(Fig. 3D), corresponding to the thermal behavior described by other authors for soy protein [6, 43-45]. In the
331
temperature range of 650 to 900 oC, further degradation of G-H 2.9% microparticles was still observed.
332
Especially, more than 18% mass loss after 600 oC was detected for the G-H 2.9% microparticles, indicating
333
that the microcapsules possessed strong heat resistance. However, the microcapsules coated by the SPI-Md
334
conjugates presented different mass loss behavior as the first degradation took place at a temperature range
335
of 100-210 oC (Fig. 3B and E). This was mainly related to the loss of the fish oil on the surface or from the
336
cracks of the microcapsules, since no degradation of the wall material was observed. The main mass loss
337
was found at the temperature range of 210-500 oC, which was attributable not only to fish oil thermal
338
degradation, but also to the initial degradation of SPI-Md conjugate. The microcapsules using the SPI/Md
339
mixture as wall material formed a broad peak ranging from 100 to 500 oC (Fig. 3C and F). Overall, the
340
microcapsule prepared by the SPI-Md conjugates and SPI/Md mixture both showed the core material
341
degradation below 200 oC, indicating poor thermal stability and inefficient microencapsulation, which was
342
consistent with the results of microencapsulation efficiency (Table 4). Nevertheless, the advantage of the
343
microcapsule of SPI-Md conjugate hydrolysate (DH 2.9%) in the thermal stability was apparently due to the
344
increased emulsifying property after limited hydrolysis, thus completely coating the fish oil droplets.
345
3.2.4 Scanning Electron Microscope (SEM) micrograph
Ac ce p
te
d
M
an
us
cr
ip t
327
346
Freeze drying was used to prepare the fish oil microcapsules. First, the emulsions were frozen, followed
347
by the removal of water by sublimation. The SEM micrographs (Fig. 4A–F) highlighted the effects of
348
limited hydrolysis of the wall material on the surface morphology of the particles. After freeze-drying
349
images like wood chips or flakes were observed, confirming the morphology previously described by other 17
Page 17 of 33
authors [46, 47]. For the fish oil droplets microencapsulated by the hydrolyzed SPI-Md conjugates (DH,
351
1.8%-4.6%) (Fig. 4A-C), a relatively smooth surface as well as a rather porous and uniform structure was
352
observed. Whereas, the extensive hydrolysis (DH, 5.7%-6.2%) (Fig. 4D-E) resulted in the particles with a
353
rougher surface, accompanied by irregular agglomerates. Moreover, similar rough morphology was
354
observed for the control sample (fish oil particles prepared with SPI-Md conjugates) as shown in Fig. 4F.
355
The porous surfaces observed were possibly formed by cavities left from ice crystals or air bubbles retained
356
during the freezing [46]. This phenomenon can be explained by the fact that better emulsification of the
357
modified SPI products after limited hydrolysis led to the stable emulsions with small oil droplets uniformly
358
coated by the wall material in the water-phase, which was just the resource of the ice crystals, and thus
359
porous and uniform structure formed after the freeze-drying process.
360
Conclusions
d
M
an
us
cr
ip t
350
This study is related to the use of hydrolyzed protein-polysaccharide conjugates in microencapsulation of
362
fish oil by freeze-drying method. Structural modifications of the native SPI were carried out by limited
363
hydrolysis following the glycosylation with maltodextrin to improve its functional properties for the
364
application as an encapsulating agent. The surface and interfacial tension decreased after limited hydrolysis
365
indicating a better amphiphilic property, which was further confirmed by the decrease of emulsion droplet
366
size as well as the PDI value. Emulsions showed shear-thinning behavior of the emulsions, and no creaming
367
or phase separation was observed after three-week storage. Additionally, the storage stability experiments
368
showed that, the oxidative stabilities (POV and headspace propanal) of microcapsule particles prepared by
369
the modified SPI products with limited hydrolysis were apparently improved. However, extensive hydrolysis
370
might generate shorter proteinic chain length, concomitantly, decreasing the MEE. The knowledge reported
371
in this study would be useful in the development of new delivery systems for bioactive substances.
Ac ce p
te
361
18
Page 18 of 33
372
373
Acknowledgments
This research was financially supported by projects of the National 125 Program of China (2011BAD23B04, 2012BAD33B05 and 2013AA102204).
375
References
376
[1] Y. Kagami, S. Sugimura, N. Fujishima, K. Matsuda, T. Kometani and
377
(2003) 2248.
378
[2] M. Oarada,
379
[3] K. Heinzelmann, K. Franke, J. Velasco and G. Márquez-Ruiz, Eur Food Res Technol, 211 (2000) 234.
380
[4] M.-Y. Baik, E.L. Suhendro, W.W. Nawar, D.J. McClements, E.A. Decker and
381
Chem Soc, 81 (2004) 355.
382
[5] S. Drusch, Y. Serfert, A. Van Den Heuvel and
383
[6] A. Nesterenko, I. Alric, F. Silvestre and
384
[7] M. Akhtar and
385
[8] J.K. Rusli, L. Sanguansri and
386
[9] L.S.R. Sanguansri, Werribee, VIC 3030, AU) and
387
AU) in, Commonwealth Scientific and Industrial Research Organisation (Limestone Avenue, Campbell,
388
ACT 2612, AU) 2012.
389
[10] M.A. Augustin, L. Sanguansri and
390
[11] C. Chung, L. Sanguansri and
391
[12] Y. Kagami, S. Sugimura, N. Fujishima, K. Matsuda, T. Kometani and Y. Matsumura2, J Food Sci,
392
68 (2003) 2248.
cr
ip t
374
P. Chinachoti, J Am Oil
M
an
J Am Oil Chem Soc, 39 (1990) 373.
d
K. Schwarz,
te
V. Durrieu,
Food Res Int, 39 (2006) 807.
Food Res Int, 48 (2012) 387.
"Colloids Surf, B ", 31 (2003) 125.
Ac ce p
E. Dickinson,
J Food Sci, 68
us
Y. Matsumura,
M.A. Augustin,
J Am Oil Chem Soc, 83 (2006) 965. M.A.C.C. Augustin, Wheelers Hill, Victoria 3150,
O. Bode2,
J Food Sci, 71 (2006) E25.
M.A. Augustin,
Food Biophys 3(2008) 140.
19
Page 19 of 33
393
[13] M. Akhtar and
E. Dickinson,
Food Hydrocoll, 21 (2007) 607.
394
[14] S. Drusch, S. Berg, M. Scampicchio, Y. Serfert, V. Somoza, S. Mannino and
395
Hydrocoll, 23 (2009) 942.
396
[15] R. Amarowicz,
397
[16] A. Karayannidou, E. Makri, E. Papalamprou, G. Doxastakis, I. Vaintraub, N. Lapteva and
398
Food Chem, 104 (2007) 1728.
399
[17] X. Guan, H. Yao, Z. Chen, L. Shan and
400
[18] Q. Zhao, H. Xiong, C. Selomulya, X.D. Chen, H. Zhong, S. Wang, W. Sun and
401
Chem, 134 (2012) 1360.
402
[19] J. O'Regan and
403
[20] J.B. Zhang, N.N. Wu, X.Q. Yang, X.T. He and
404
[21] S. Petruccelli and
405
[22] H. Frister, H. Meisel and
406
631.
407
[23] J. Adler-Nissen, Elsevier Applied Science Publishers, 1986.
408
[24] R. Liang, S. Xu, C.F. Shoemaker, Y. Li, F. Zhong and
409
7548.
410
[25] A. Millqvist-Fureby,
411
[26] U. Klinkesorn, P. Sophanodora, P. Chinachoti, E.A. Decker and
412
(2006) 449.
413
[27] H. W, in:
414
ed, AOAC International Gaithersberg, MD, 2002.
415
[28] A. Nesterenko, I. Alric, F. Violleau, F. Silvestre and
K. Schwarz, Food
Eur J Lipid Sci Tech, 112 (2010) 695.
Food Chem, 101 (2007) 163.
M.C. Anon,
Food Chem, 123 (2010) 21. L.J. Wang,
Food Hydrocoll, 28 (2012) 301.
M
D.M. Mulvihill,
Q. Zhou, Food
an
us
M. Zhang,
cr
ip t
G. Articov,
J Agric Food Chem, 43 (1995) 2001.
Ac ce p
te
d
E. Schlimme, Fresenius Zeitschrift fur Analytische Chemie, 330 (1988)
Q. Huang, J Agric Food Chem, 60 (2012)
"Colloids Surf, B ", 31 (2003) 65. D.J. McClements,
Food Res Int, 39
Peroxide value of oils and fats, Official methods of analysis of AOAC international, 17th
V. Durrieu,
Food Res Int, 53 (2013) 115.
20
Page 20 of 33
416
[29] Y. Zhang, C. Tan, X. Zhang, S. Xia, C. Jia, K. Eric, S. Abbas, B. Feng and
417
Technol, (2014) 10.1007/s00217.
418
[30] L. Chen, J. Chen, J. Ren and
419
[31] M.P. Rascón, C.I. Beristain, H.S. García and
420
[32] K.D. Martínez, C. Carrera Sánchez, J.M. Rodríguez Patino and
421
(2009) 2149.
422
[33] M. Corzo-Martínez, C. Carrera Sánchez, F.J. Moreno, J.M. Rodríguez Patino and
423
Food Hydrocoll, 27 (2012) 438.
424
[34] M. Corzo-Martínez, C. Carrera-Sánchez, M. Villamiel, J.M. Rodríguez-Patino and
425
Food Eng, 113 (2012) 461.
426
[35] J. Maldonado-Valderrama and
J.M.R. Patino,
427
[36] Y. Wang, D. Li, L.-J. Wang and
B. Adhikari,
428
[37] R. Pal,
429
[38] P. Fäldt and
430
[39] L. Atarés, J. Bonilla and
431
[40] A.I. Bourbon, A.C. Pinheiro, M.A. Cerqueira, C.M.R. Rocha, M.C. Avides, M.A.C. Quintas and
432
Vicente,
433
[41] A. Soottitantawat, H. Yoshii, T. Furuta, M. Ohkawara and
434
[42] N. Mongenot, S. Charrier and
435
[43] H. Wang, L. Jiang and
436
[44] V. Schmidt, C. Giacomelli and
437
[45] P. Guerrero, A. Retegi, N. Gabilondo and
438
[46] M.C. Farias, M.L. Moura, L. Andrade and
Eur Food Res
Food Hydrocoll, 25 (2011) 887. Food Sci Technol Res , 44 (2011) 549. A.M.R. Pilosof, Food Hydrocoll, 23
ip t
M.A. Salgado,
M. Villamiel,
F.J. Moreno,
J
an
us
cr
M. Zhao,
F. Zhong,
M
Curr Opin Colloid Interface Sci, 15 (2010) 271. J Food Eng, 104 (2011) 56.
Food Sci Technol Res, 29 (1996) 438.
te
B. Bergenståhl,
d
Curr Opin Colloid Interface Sci, 16 (2011) 41.
J Food Eng, 100 (2010) 678.
Ac ce p
A. Chiralt,
A.A.
J Food Eng, 106 (2011) 111.
L. Fu,
a.P. Chalier,
P. Linko,
J Food Sci, 68 (2003) 2256.
J Agric Food Chem, 48 (2000) 861.
J Appl Polym Sci, 106 (2007) 3716. V. Soldi,
Polym Degrad Stab, 87 (2005) 25. K. de la Caba,
J Food Eng, 100 (2010) 145.
M.H.M.R. Leão,
Mater Res, 10 (2007) 57.
21
Page 21 of 33
439
[47] M. Sousdaleff, M.L. Baesso, A.M. Neto, A.C. Nogueira, V.A. Marcolino and
440
Food Chem, 61 (2013) 955.
G. Matioli, J Agric
441 442
Ac ce p
te
d
M
an
us
cr
ip t
443
22
Page 22 of 33
443
Figure captions
444
Fig. 1 Surface (A) and interfacial (B) tension of combined modified SPI products and native SPI as a
446
function of concentration. The points in the graphic are means ± standard deviation of three separate
447
determinations (mean± SD, n = 3).
448
Fig. 2 Effects of different degree of hydrolysis (DH) of combined modified SPI products on the rheological
449
properties of emulsions with fish oil/protein ratio of 0.67.
450
Fig. 3 The thermogravimetric weight loss (TG) (A, B, C) and derivative of weight loss (DTG) (D, E, F)
451
curve of the fish oil microparticles stabilized by the modified SPI products.
452
Fig. 4 Scanning electron micrographs of freeze dried fish oil microparticles stabilized by various combined
453
modified SPI products (A-E: DH 1.8%, DH 2.9%, DH 4.6%, DH 5.7%, DH 6.2%; E: SPI-Md).
Ac ce p
te
d
M
an
us
cr
ip t
445
23
Page 23 of 33
*Graphical Abstract (for review)
Ac
ce pt
ed
M
an
us
cr
ip t
Graphical Abstract
Page 24 of 33
*Highlights (for review)
Highlights Wall material of microcapsules is crucial for protecting fish oil from oxidation.
Hydrolysate of SPI-Maltodextrin conjugate gives better stabilization.
Degree of hydrolysis strongly affects the emulsions and powders of fish oil.
Emulsions coated by limited hydrolyzed products showed better emulsion stability.
Powders coated by hydrolysates (DH 2.9%) showed better oxidative stability.
Ac
ce pt
ed
M
an
us
cr
ip t
Page 25 of 33
Table(s)
Tables
Table 1 The total solids, viscosity, particle size and size distribution of combined modification SPI/fish oil emulsion after high pressure homogenizationa
7.51±0.06 8.00±0.26a 8.30±0.17a 10.90±0.27b 12.20±0.55c
a
3.69±0.05 3.75±0.06a 4.63±0.07c 4.09±0.17b 4.91±0.05c
Z-average (nm) b
273.0±2.52 317.7±4.39c 235.5±0.29a 282.1±1.16b 407.4±5.20d
PDI 0.315±0.002c 0.204±0.010b 0.162±0.001a 0.203±0.005b 0.374±0.007d
The emulsions stabilized by combined modified SPI products (DH 5.7%), with two-stage homogenization (35+10
us
a
a
Viscosity (mPas)
ip t
0.25 0.43 0.67 1.0 1.5
Total solids (% w/w)
cr
Oil/Protein
Ac
ce pt
ed
M
an
MPa). All emulsions were prepared in triplicate. Different lower case letters are significantly different (P < 0.05).
Page 26 of 33
Table 2 The total solids, viscosity, particle size and size distribution of combined modified SPI/fish oil emulsiona initially prepared and 3 weeks (4 oC) after high pressure homogenization
a
9.90±0.12c 11.4±0.58d 8.90±0.26ab 8.30±0.23a 9.70±0.12bc 8.55±0.29a 9.81±0.12bc
4.05±0.06a 5.32±0.08c 4.00±0.29a 4.63±0.18b 5.10±0.17c 3.56±0.12a 4.61±0.12b
Initially prepared Z-average (nm) PDI a 212.5±1.01 0.115±0.001ab 232.6±1.85b 0.091±0.021a d 329.3±2.37 0.193±0.023c 253.6±2.92c 0.154±0.011b 254.8±0.27c 0.116±0.004ab 401.3±4.61f 0.293±0.010d 344.5±4.32e 0.279±0.004d
After 3 weeks storage Z-average (nm) PDI a 251.3±1.12 0.166±0.005a 255.6±1.15a 0.154±0.006a b 266.6±1.15 0.167±0.004a 281.5±0.58c 0.157±0.004a 288.5±1.01c 0.156±0.006a 585.2±5.85e 0.490±0.011b 491.6±5.60d 0.419±0.003c
ip t
Viscosity (mPas)
cr
Sample 1.80% 2.90% 4.60% 5.70% 6.20% SPI/Md(2/1)a SPI-Md(2/1)b
Total solids (% w/w)
The emulsions were prepared with the oil/protein ratio of 0.67/1 and two-stage homogenization (35+10 MPa). All
SPI-Md (2/1) was the conjugates of the SPI and Maltodextrin at 80 oC for 80 min with the ratio of 2/1. Different
an
c
us
emulsions were prepared in duplicate. b SPI/Md (2/1) was the mixture of SPI and Maltodextrin with the ratio of 2/1.
Ac
ce pt
ed
M
lower case letters are significantly different (P < 0.05).
Page 27 of 33
Table 3 Rheological parameters obtained from fitting using Power Law Model for fish oil emulsions prepared with different modified SPI products K (Pa·sn)
n
R2
1.8% 2.9% 4.6% 5.7% 6.2%
0.011±0.001 0.014±0.001 0.010±0.001 0.013±0.001 0.017±0.001
0.831±0.014 0.825±0.010 0.833±0.014 0.810±0.011 0.773±0.007
0.997 0.998 0.996 0.998 0.999
Ac
ce pt
ed
M
an
us
cr
ip t
Sample
Page 28 of 33
Table 4 The MEE and oxidative stability (headspace propanal and POV)a of the combined modification SPI/fish oil microcapsules Initially prepared Sample
MEE (%)
4 Weeks
Headspace POV
propanal
(mmol/kg)
8 Weeks
Headspace Retention
POV
time(min)
(mmol/kg)
(GC area)
propanal
Headspace Retention
POV
time(min)
(mmol/kg)
(GC area)
Retention time(min)
(GC area)
97.84
1.10
1050.98
1.057
9.77
5750.49
1.058
12.60
8728.72
1.034
2.90%
95.08
1.77
537.60
1.055
4.72
2688.00
1.054
8.06
2980.59
1.040
4.60%
82.22
1.66
1301.84
1.060
12.90
5069.18
1.060
5.70%
90.52
1.99
1732.73
1.058
17.45
6863.65
1.057
91.47
1.99
1421.02
1.061
24.54
7105.09
1.061
66.31
2.42
2861.60
1.052
24.06
10848.27
1.053
SPI-Md (2/1)
26.93
2.36
2909.38
1.054
24.06
10496.89
1.054
16.67
9646.57
1.042
19.32
9191.57
1.037
27.79
9040.13
1.042
27.82
16691.58
1.041
28.11
20306.12
1.040
cr
6.20%
ip t
1.80%
SPI/Md (2/1)
ce pt
ed
M
an
us
All the values shown in the table were the means of three parallel experiments (n=3).
Ac
a
propanal
Page 29 of 33
Figure(s)
Figure
Ac
ce pt
ed
M
an
us
cr
ip t
Fig. 1
Page 30 of 33
Ac
ce pt
ed
M
an
us
cr
ip t
Fig. 2
Page 31 of 33
Ac
ce pt
ed
M
an
us
cr
ip t
Fig. 3
Page 32 of 33
Ac
ce pt
ed
M
an
us
cr
ip t
Fig. 4
Page 33 of 33
S0927-7765(14)00232-X http://dx.doi.org/doi:10.1016/j.colsurfb.2014.05.006 COLSUB 6411
To appear in:
Colloids and Surfaces B: Biointerfaces
Received date: Revised date: Accepted date:
24-2-2014 28-4-2014 2-5-2014
Please cite this article as: Y. Zhang, C. Tan, S. Abbas, K. Eric, X. Zhang, S. Xia, C. Jia, The effect of soy protein structural modification on emulsion properties and oxidative stability of fish oil microcapsules, Colloids and Surfaces B: Biointerfaces (2014), http://dx.doi.org/10.1016/j.colsurfb.2014.05.006 This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to our customers we are providing this early version of the manuscript. The manuscript will undergo copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please note that during the production process errors may be discovered which could affect the content, and all legal disclaimers that apply to the journal pertain.
1
The effect of soy protein structural modification on emulsion properties and
2
oxidative stability of fish oil microcapsules Yating Zhang, Chen Tan, Shabbar Abbas, Karangwa Eric, Xiaoming Zhang *, Shuqin Xia, Chengsheng Jia
4
State Key Laboratory of Food Science and Technology, School of Food Science and Technology, Jiangnan
5
University, Lihu Road 1800, Wuxi, Jiangsu 214122, China.
us
cr
ip t
3
* Corresponding author: Tel: +86 510 85197217; Fax: +86 510 85884496.
7
E-mail address: [email protected] (X. Zhang)
Ac ce p
te
d
M
an
6
1
Page 1 of 33
8
9
Abstract
Hydrolysates of soy protein isolate-maltodextrin (SPI-Md) conjugate were used as wall material to
11
prepare fish oil microcapsules by freeze-drying method. Effects of the protein structural modifications on the
12
physicochemical properties of the emulsion and the oxidative stability of the microcapsules were
13
characterized. Compared with emulsions of SPI-Md conjugates or soy protein isolate/maltodextrin (SPI/Md)
14
mixture, lower droplet size (212.5-329.3 nm) and polydispersity index (PDI) (0.091-0.193) were obtained in
15
the fish oil emulsions prepared by SPI-Md conjugate hydrolysates. The improved amphiphilic property of
16
SPI-Md conjugate hydrolysates was supported by the results of surface and interfacial tension, and further
17
confirmed by the improved emulsion stability during the storage period. Although the microencapsulation
18
efficiency (MEE) of SPI-Md conjugate hydrolysates slightly decreased from 97.84% to 91.47% with the
19
increasing degree of hydrolysis (DH), their oxidative stabilities (Peroxide value and headspace propanal)
20
were apparently improved compared with native SPI/Md mixture or SPI-Md conjugates system. Moreover,
21
favourable thermal stability as well as a porous and uniform surface structure of the microcapsules coated by
22
SPI-Md conjugate hydrolysates (DH 2.9%) was observed via the thermal analysis and scanning electron
23
microscope (SEM) micrographs, respectively.
24
Key words
26
cr
us
an
M
d
te
Ac ce p
25
ip t
10
Microcapsules; Soy protein isolate; Conjugates; Limited hydrolysis
1. Introduction
27
Fish oil is known to be a rich source of long chain polyunsaturated fatty acids such as the omega-3 (n-3)
28
family, and many health benefits have been associated with their regular consumption. However, n-3 fatty 2
Page 2 of 33
29
acids are susceptible to oxidative deterioration, which restricts their use in foods [1]. Especially,
30
hydroperoxides, the primary product of lipid oxidation have been considered to be toxic [2].
31
Microencapsulation was reported to be a desirable technique for preventing the oxidation of n-3 fatty acids
32
[1, 3-5]. The selection of wall materials, which significantly affected the stability of powders, was mainly based on
34
their interfacial functionality [6]. The water solubility and amphiphilic properties, the ability to self-associate
35
and interact with variety of substances, and the high molecular chain flexibility of modified proteins were all
36
proved conducive to emulsification [6]. Vegetable proteins are relatively cheap source of wall material for
37
active component microencapsulation, besides their nutritional value and biodegradability. As there is
38
pressing need for multi-functional materials, different modification techniques have been developed to
39
enhance and diversify the protein functionalities, in order to make them be more suitable wall materials for
40
the current microencapsulation techniques.
M
an
us
cr
ip t
33
It was reported that the glycosylated proteins had good emulsifying [7] as well as antioxidant properties
42
[8]. Augustin research group [9-12] confirmed that glycosylated proteins could be used as effective
43
encapsulating materials in the microencapsulation of oils. Akhtar and Dickinson [13] demonstrated that the
44
whey protein–maltodextrin conjugates possessed excellent emulsifying properties and could be used as an
45
alternative to gum arabic. Recently, Drusch et al. [14] reported that the Maillard reaction in aqueous solution
46
could lead to an increase in redox-active compounds of the caseinate–glucose syrup, thus increasing the
47
oxidative stability of encapsulated fish oil. Moreover, Ryszard Amarowicz [15] summarized that enzymatic
48
hydrolysis of proteins was another powerful tool in the modification of their functional properties in food
49
systems. Limited hydrolyzed proteins were reported to exhibit improved emulsifying, solubility and higher
50
antioxidant activities than that of the non-hydrolyzed protein, due to the stretched protein chains [16-18].
51
Nevertheless, single modification method may not be enough to fulfill the pressing demand for
Ac ce p
te
d
41
3
Page 3 of 33
multi-functional materials. Recently, a novel method of protein modification through the limited enzymatic
53
hydrolysis of protein-polysaccharide conjugates has attracted considerable interest. For instance, the sodium
54
caseinate-maltodextrin conjugate hydrolysates were reported to have potential as a water soluble low
55
molecular weight emulsifier and an emulsion stabilizer [19]. Similar results were reported by Zhang et al.
56
[20], which confirmed that the combination of protein-polysaccharide conjugation and controlled enzymatic
57
hydrolysis was a useful strategy to prepare an effective emulsifier.
cr
ip t
52
However, few studies mentioned the application of these hydrolyzed protein-polysaccharide conjugates as
59
wall materials for microencapsulation, especially for fish oil. This study was conducted to evaluate the effect
60
of SPI modifications on the emulsion properties and the oxidative stability of microencapsulated fish oil.
61
Limited enzymatic hydrolysis was carried out following the preparation of soy protein isolate-maltodextrin
62
conjugates through Maillard reaction. Freeze drying was selected mainly due to its lower drying temperature,
63
which could help to protect the fish oil from the oxidative deterioration. It was expected to get further
64
insights into the structural modifications of wall material and its influence on the microencapsulation
65
properties, thus, offering more reference information for proteins modification in the future.
66
2. Materials and Methods
67
2.1 Materials
Ac ce p
te
d
M
an
us
58
68
The defatted soybean meal was obtained from Anyang Mantianxue Food Manufacturing Company
69
(Henan, China). Maltodextrin (Md) with DE values of 8-10, was obtained from Baolingbao Biology
70
Company (Shangdong, China). The enzyme Neutrase was obtained from Novozymes (Jiangsu, China). The
71
fish oil contained approximately 70% EPA/DHA (20/50). All other chemicals used in this study were of
72
analytical grade.
4
Page 4 of 33
73
2.2 Preparation of the Soy Protein Isolate (SPI) SPI was prepared according to the method of Petruccelli and Anon [21]. The defatted soybean meal
75
powder was suspended in 15-fold water and adjusted to pH 7 with 2M NaOH. After stirring for 1 h, the
76
suspension was centrifuged at 8000 g for 30 min and the supernatant was subjected to isoelectric
77
precipitation by adjusting pH to 4.5 with 2M HCl. The protein precipitate, which was recovered by
78
centrifuging (8000 g, 30 min), was re-suspended in water and adjusted to pH 7 with 2M NaOH. After
79
removing small amount of insoluble substances by centrifuging at 8000 g for 30 min, the protein solution
80
was freeze dried and ground to yield SPI powder. All procedures were carried out at room
81
temperature. Crude protein content of SPI determined by Kjeldhal method was 96.7% (w/w) (N×6. 25).
82
2.3 Hydrolysis after glycosylation
M
an
us
cr
ip t
74
SPI (4%, w/v, protein) was incubated with maltodextrin at the SPI/maltodextrin ratio of 2/1 (w/w) for 80
84
min at 80 oC, pH 7.0, and the degree of glycosylation (DG) was 33% determined according to the modified
85
o-phthaldialdehyde (OPA) method [22].
86
The temperature of the glycosylated solution was raised to 54 oC, pH 7.0, followed by enzymatic hydrolysis
87
with Neutrase (Enzyme/Substance (E/S) = 0.5%, 2%, 4%, 6%, 8%) for 25 min to DH 1.8%, DH 2.9%, DH
88
4.6%, DH 5.7%, DH 6.2%, respectively, through pH-stat method [23]. The progress of the reaction process
89
was monitored by the consumption of 0.5 M NaOH. Afterwards, the enzyme was inactivated by heating at
90
80 oC for 20 min. These samples were lyophilized and referred as SPI-Md conjugate hydrolysates.
91
2.4 Emulsion preparation
Ac ce p
te
d
83
92
The modified products of SPI were initially dispersed in warm (70 °C) distilled water. The pH of the
93
resulting solution was adjusted to 7.0 with 1 M NaOH. The oil phase was heated to 70 °C in a water bath
94
prior to coarse emulsion preparation through homogenizer (ULTRA-TURRAX T-25, IKA, Staufen, 5
Page 5 of 33
95
Germany) at 10,000 rpm for 2 min. The coarse emulsion was subjected to two-stage homogenization (35+10
96
MPa) using a high pressure homogenizer (NS1001 L2K, A S Co. Ltd., Parma, Italy).
97
2.5 Emulsion size The particle size distribution of the emulsion droplets was measured by using a Malvern Mastersizer 2000
99
(Malvern Instruments Ltd., Malvern, England). The emulsion was diluted with distilled water. A relative
100
refractive index ηoil/ηwater = 1.095 (ηoil = 1.465 and ηwater = 1.330) was used for the calculation of particle
101
size distribution, assuming that all droplets were spherical in shape. A polydisperse model was used to
102
analyze the data while the intensity of scattered light was taken as an indication of particle size.
103
2.6 Surface and interfacial tension
an
us
cr
ip t
98
The influence of different concentrations of the modified SPI products on the interfacial tension at the fish
105
oil-water phase and the air- water surface tension were measured by the Wilhelmy plate method with an
106
automated tensiometer DCAT21 (DataPhysics instrument GmbH, Germany) at 25 oC. The sample
107
concentration varied between 0.5% and 4.0% (w/w) at pH 7.0.
108
2.7 Rheological measurements of emulsions
Ac ce p
te
d
M
104
109
Rheological measurements of fish oil emulsions were performed at 25 °C according to the method of
110
Liang et al. [24] using the Advanced Rheometric Expansion System (ARES, TA Instruments, New Castle,
111
DE, USA) with a cone and plate geometry (cone diameter = 50 mm, angle = 4°, gap = 0.05 mm). A thin
112
layer of silicone oil was applied on the surface of the samples in order to prevent evaporation. For each
113
measurement, 1.5 mL of the emulsion sample was loaded on the rheometer. The viscosity of nanoemulsions
114
was measured by a steady state flow program with the shear rate ranging from 0 s−1 till steady. Experimental
115
flow curves were fitted to a power law model 6
Page 6 of 33
116
η = Kγn-1
(1)
where η was the viscosity (Pa·s), γ was the shear rate (s−1), K was the consistency index (Pa·sn), and n was
118
the index that provided information about the flow behavior related to the effect of shear rate. There exist
119
three value ranges for n: n < 1 for a shear-thinning fluid, n = 1 for a Newtonian fluid, and n > 1 for a
120
shear-thickening fluid.
121
2.8 Preparation of the Microcapsule powders
us
cr
ip t
117
Emulsions were frozen with liquid nitrogen and subjected to a freeze-drying process (Labconco FreeZone
123
4.5L, freeze-drier, Kansas City, MO, USA). The operating condition of freeze-dryer was −50 °C for a period
124
of up to 24 h for all samples. After the drying process, a fine dry powder was obtained.
125
2.9 Extraction of free oil
d
M
an
122
Extraction of the free oil was made with petroleum ether (b.p. 60-90 oC) following the method as
127
described by Anna Millqvist-Fureby (2003) [25], which was modified to suit the sample volume. One gram
128
of powder was added to 10 ml dried petroleum ether, and shaken for 2 min. Solvent was then separated by
129
filtration. The solid residue was washed with 2×2 ml petroleum ether. The combined filtrate was evaporated
130
using a rotary evaporator. Fat residue was then dried at 105 oC until a constant weight was reached.
131
2.10 Extraction of total oil
Ac ce p
te
126
132
The extraction of total oil was based on the method of Utai Klinkesorn et al. with some modification [26].
133
Five-milliliters of distilled water (60 oC) was added to 0.5 g powder and shaken for 15 min using thermostat
134
Water Bath Vibrator (CD3192, Xutemp, Hangzhou, China). The resulting solution was then extracted with
135
25 mL hexane/isopropanol (3:1, v/v). The tubes were then vortexed for 15 min, and centrifuged for another 7
Page 7 of 33
15 min at 8000 g. The clear organic phase was collected while the aqueous phase was re-extracted with the
137
solvent mixture [3, 4]. After filtration through anhydrous Na2SO4, solvent was evaporated in a rotary
138
evaporator (RE-52AA, Shanghai Biochemical Instrument Company, China) at 70 oC. The solvent-free
139
extract was dried at 105 oC. The amount of encapsulated oil was determined gravimetrically.
140
2.11 Microencapsulation efficiency (MEE)
The encapsulation efficiency (EE) was calculated as follows:
142
EE (%) = (Total oil - free oil)/Total oil×100
an
(2)
2.12 Powder storage
M
143
us
141
cr
ip t
136
Five grams of powder was placed in 20 mL loosely capped amber glass bottle. Samples were stored at 35
145
°C for 4 and 8 weeks in the dark. The extent of lipid oxidation in the powder was monitored by measuring
146
the Peroxide value (POV) and headspace propanal.
147
2.13 Peroxide value (POV)
Ac ce p
te
d
144
148
Hydroperoxides, the primary oxidation products, were measured according to AOAC Official Method
149
965.33 [27], with small modification to suit small sample volumes. 15 mL CH3COOH–CHCl3 was added
150
into 0.50 ± 0.05 g test portion, and the mixture was stirred well to destroy the encapsulants. Then 0.05 mL
151
saturated KI solution was added from Thermal pipet, followed by occasional shaking 1 min, and 30 mL H2O
152
was added. It was slowly titrated with 0.01M Na2S2O3 with vigorous shaking until yellow disappear.
153
Subsequently, 0.05 mL 1% starch solution was added, titration and vigorously shaking was continued to
154
release all I2 fromCHCl3 layer, until blue just disappeared.
155
Peroxide value (milliequivalent peroxide/kg oil or fat) = S×M×1000/g sample, where S = mL, Na2S2O3 8
Page 8 of 33
156
(blank corrected); M = molarity, Na2S2O3 solution.
157
2.14 Headspace propanal A PerkinElmer Model Autosystem XL capillary gas chromatograph (GC) fitted with a DB1 fusedsilica
159
capillary column (25 m × 0.32 mm, 5 µm film thickness; Agilent Technologies, Forest Hill, Victoria,
160
Australia) and an flame
161
the powder sample was weighed and sealed in a 20 mL headspace vial before being equilibrated at 60 °C for
162
15 min. Approximately 2.5 mL of the headspace vapor was injected into the column. The column
163
temperature was increased initially from 60 °C to 75 °C at the rate of 3 °C/min, then to 90 °C at the rate of 5
164
°C/min and finally to 230 °C at the rate of 25 °C/min, where it was held for 20 min. The detector
165
temperature was 240 °C. Headspace propanal content was reported as absolute GC area.
166
2.15. Thermal analysis
ip t
158
d
M
an
us
cr
ionization detector (FID) was used for propanal headspace analysis. One gram of
The thermogravimetric curves (TG) were obtained with the thermal analysis system Q500IR (TA
168
Instruments, USA). Samples were heated from 30 to 900 oC using open aluminium crucibles with
169
approximately 6.0 mg of the sample under a synthetic air flow of 150 mL min-1 at a heating rate of 10 oC
170
min-1. The instrument was preliminarily calibrated with standard mass and with standard calcium oxalate
171
monohydrate. The derivative thermogravimetric curves (DTG), the first derivative of TG curves, were
172
calculated.
173
2.16. Scanning Electron Microscopy (SEM) Measurement.
Ac ce p
te
167
174
Scanning electron micrographs of freeze-dried samples were recorded with a Quanta-200 scanning
175
electron microscope (FEI Co., Eindhoven, The Netherlands) at an accelerating voltage of 10 kV. The
176
powders were sprinkled onto double-backed cellophane tape attached to an aluminum stub before coating 9
Page 9 of 33
177
with gold-palladium in an argon atmosphere.
178
2.17. Statistical analysis The experiments were performed in triplicate and values were expressed as mean ± standard error (SD).
180
All data were subjected to analysis of variance (ANOVA), and the differences between means were
181
evaluated by Duncan’s multiple range test.
182
3. Results and Discussion
183
3.1 Characterization of emulsions prior to drying.
184
3.1.1 Effect of fish oil/protein ratio on the preparation of emulsion
M
an
us
cr
ip t
179
For oil/water (O/W) emulsions, it is always desirable to contain high core volume in order to increase the
186
concentration of hydrophobic active ingredients. Whereas, fish oil content can also affect the emulsion
187
droplet sizes through a coalescence phenomenon [24]. Thus, the fish oil/protein is an important parameter to
188
evaluate the stability of O/W emulsions produced.
Ac ce p
te
d
185
189
Emulsions with various fish oil/protein ratios stabilized by combined modified SPI products (DH 5.7%)
190
were prepared. Formulations and processing conditions for the emulsion preparation are shown in Table 1.
191
The increase of fish oil/protein ratio resulted in a progressive increase of the total solids content, which was
192
the main factor responsible for viscosity increase. However, the change of viscosity had no effect on the
193
preparation of the emulsions as the viscosity was still low. The Z-average diameter and size distributions of
194
emulsions prepared by different fish oil/protein ratios are presented in Table 1. The particle size was slightly
195
changed when the fish oil/protein ratio was in the range of 0.25-1. However, when the fish oil/protein ratio 10
Page 10 of 33
reached to 1.5, a clear increase of the particle size from 282.1 nm to 407.4 nm was observed due to the
197
coalescence [28]. In this case, the excessive oil volume fraction could not be covered by the emulsifier, and
198
part of the free oil droplets would affect the droplet diameter of the emulsion [24].
ip t
196
The polydispersity index (PDI) value represents the particle size distribution of the droplets. A small PDI
200
value indicates a narrow particle size distribution [24]. According to the results shown in Table 1, the PDI
201
values for emulsions stabilized by the modified SPI of DH 5.7% were normally < 0.4, indicating that all of
202
the emulsions had a relatively narrow range of size distribution. The increasing trend of the PDI was
203
apparent, as the oil/protein ratio increased from 0.67 to 1.0. Emulsions showed a small particle size (235.5
204
nm) and a narrow size distribution (PDI, 0.162) at oil to the protein mass ratio of 0.67.
205
3.1.2 Influence of combined modifications on the emulsion properties
M
an
us
cr
199
One of the most important factors leading to efficient microencapsulation of active compounds is their
207
effective dispersion in the wall material solution [28]. The retention of active material during drying process
208
was reported to be increased with the decrease of mean emulsion oil droplet size [8].
Ac ce p
te
d
206
209
The Z-average diameter and PDI of emulsions stabilized by different modified SPI products with the
210
oil/protein ratio of 0.67 are shown in Table 2. It can be seen that the Z-average diameter and PDI of
211
emulsions showed a significant (P < 0.05) increase with the increase of DH. This can be explained by the
212
fact that although limited enzymatic hydrolysis could help to expose the hydrophobic groups and enhance
213
the amphiphilic characteristics, further hydrolysis resulted in shorter peptides, which could affect on the
214
formation of the bulkier polymeric layer around the oil droplets and the stabilization of the oil emulsion [29].
215
Overall, emulsion stabilized by the hydrolyzed conjugates of DH 2.9% displayed desirable emulsifying
216
properties. This observation was consistent with the results reported by Chen et al., which stated that 11
Page 11 of 33
217
excessive hydrolysis may destroy the steric structure of the amphiphilic molecule, which contributed to the
218
emulsifying properties [30]. However, the Z-average diameter values are smaller than those from previous studies on native vegetable
220
proteins used as wall material for microencapsulation [6, 8]. The more efficient dispersion behavior of fish
221
oil (resulting in smaller droplets) can be explained by the presence of stretch proteinic chains after
222
hydrolysis or by improved protein active surface properties after glycosylation, which resulted in the
223
enhancement of the molecular flexibility [6]. Small emulsion Z-average diameter resists against coalescence
224
during the drying process and make emulsions more stable [31]. Therefore, the combined modification
225
method seemed to be efficient for improving the stability and properties of emulsion.
an
us
cr
ip t
219
Additionally, to assess the extended stability of these emulsions, 3 weeks storage test was carried out at 4
226 227
o
228
and the Z-average diameter and PDI kept consistent, indicating an improved storage stability of the
229
combined modified SPI oil-in-water emulsions (Table 2).
230
3.1.3 The surface and interfacial tension
Ac ce p
te
d
M
C. Over this period of storage, there was no obvious phase separation or creaming observed for all samples
231
Changes in the molecular structure of soy protein as a consequence of combined modification may lead to
232
specific surface characteristics or interfacial properties. Fig. 1 shows the surface tension and interfacial
233
tension of native SPI as well as the combined modified SPI products with various degree of hydrolysis (DH),
234
respectively. It was observed that both the surface and interfacial tension decreased with increasing
235
concentration, thus, reflecting better surface and interfacial properties. A slight decrease in surface tension
236
was observed (Fig. 1A) after limited hydrolysis (DH 1.8%, 2.9% and 4.6%) of the glycosylated SPI products.
237
However, the combined modified samples at higher degree of hydrolysis (DH 5.7%-6.2%) showed similar
238
surface tension with the non-hydrolyzed products. This behavior was in agreement with the results of Karina
239
et al. (2009), who pointed out that a low degree of hydrolysis (2–5%) would be enough to improve the 12
Page 12 of 33
240
surface activity of soy protein [32]. High degree of hydrolysis probably decreased the interaction ability of
241
the modified soy protein products at the air-water interface and caused a decrease in foaming ability. The
242
direct relationship between the foam formation, stability and the interfacial properties of adsorbed protein
243
films was reported by Marta research group [33, 34]. The interfacial behavior is an important aspect of both emulsion formation and stability. Proteins are
245
surface-active substances used to stabilize the emulsions by adsorbing at oil-water interface, thereby,
246
reducing the interfacial tension, and forming the mechanical energy barrier at the interface against
247
coalescence in the emulsion system [35]. As shown in Fig. 1B, there was no remarkable difference in the
248
interfacial tension of the native SPI, the glycosylated SPI products and the SPI-Md conjugates hydrolysates
249
(DH 1.8%). However, the limited hydrolysis of DH 2.9% led to a slight decrease of the interfacial tension
250
from 4.058±0.026 mN/m (the glycosylated SPI products ) to 3.354±0.026 mN/m (DH 2.9%) at the
251
equilibrium state, which may be due to the unfolding of the molecule and the increase of protein
252
hydrophobicity which provides a better adsorption process at the oil-water interface [30]. However, further
253
hydrolysis (DH 4.6%-6.2%) resulted in an obvious increase of the interfacial tension compared with the
254
non-enzymatic products. This phenomenon was corresponding to the discussion above on the emulsion
255
properties.
256
3.1.4 The rheological properties of the emulsions
Ac ce p
te
d
M
an
us
cr
ip t
244
257
The rheological properties of the emulsions stabilized by modified SPI products were also investigated to
258
evaluate their emulsion stability. Most emulsion systems used in foods and beverages show shear-thinning
259
behavior, which is important for lowering the viscosity under flow during consumption [24]. The shear rate
260
dependence of apparent viscosities of fish oil emulsions was investigated as shown in Fig. 2. Table 3 shows 13
Page 13 of 33
the rheological parameters fitted with a power law function for emulsions prepared with modified SPI
262
products. The square of the correlation index (R2) was >0.95, suggesting that the model is suitable for the
263
emulsions studied in this work. The flow behavior index (n) indicates the extent of the shear-thinning
264
behavior [36]. Similar typical shear-thinning behaviors were observed for all emulsions (Fig. 2)
265
corresponding to the decreasing tendency of the flow behavior indices (n), which were <1.0 (Table 3).
266
However, there was a slight increase of K values for the emulsions with the increasing DH of modified SPI
267
products, in agreement with the increase of viscosity (Table 2). The change of rheological properties was
268
mainly due to the change in droplet mean diameter [24]. These observations were consistent with the
269
previous report, suggesting that the emulsions of smaller sizes had much higher viscosities than the coarse
270
emulsions and the shear-thinning effect was much stronger in the case of smaller-sized emulsions [37]. As a
271
whole, these characteristics were typical for non-aggregated emulsion droplets, thus, indicating the
272
emulsions stability with no visual instability such as creaming or separation during the storage.
273
3.2 Characterization of the Microcapsule powders
Ac ce p
te
d
M
an
us
cr
ip t
261
274
The microencapsulation efficiency, oxidative stability (headspace propanal and POV), thermal stability
275
and morphology of freeze-dried powders prepared from combined modified SPI products were investigated
276
(Table 4).
277
3.2.1 Microencapsulation efficiency (MEE).
278
The microencapsulation efficiency (MEE) of the powders encapsulated by SPI-Md conjugate hydrolysates
279
was much higher than that of the powders encapsulated by the mixture of SPI and Md, or their conjugates
280
(Table 4). However, a slight decreasing tendency of MEE from 97.84% to 91.47% was observed with 14
Page 14 of 33
increasing DH from 1.8% to 6.2%. That was possibly due to the fact that the shorter chain length of wall
282
material cannot produce a sufficient strong structural matrix to encapsulate the fish oil. Proteinic chains
283
accumulate at the air/water interface of emulsion droplets during drying and thus represent the surface of
284
formed powder particles [38]. Furthermore, the oil droplet size is associated with the microencapsulation
285
efficiency, and small droplet size is expected to achieve high microencapsulation efficiency [14].
286
Additionally, increased oil droplet size in the emulsion might result in higher amount of surface oil.
287
3.2.2 Oxidative stability.
us
cr
ip t
281
To assess the oxidative stability of fish oil powders, an extended storage test (4 weeks or 8 weeks, 35 oC)
289
was carried out. The formation of the hydroperoxides and propanal were measured during the storage time
290
(Table 4).
M
an
288
The fish oil alone was highly oxidizable as the rate of lipid oxidation was very fast. After two-week
292
storage in the same condition, the contents of the POV and headspace propanal soared from 0.65 mmol/kg
293
and 127.5 initially to 157349 mmol/kg and 18757, respectively (Data not shown). However, although the
294
yield of the oxidation products increased with increasing DH, the oxidative stabilities of fish oil powders
295
encapsulated by the SPI-Md conjugate hydrolysate were generally superior to the corresponding powders
296
prepared with SPI-Md conjugates or the SPI/Md mixture, as observed by the lower peroxide value (POV)
297
and headspace propanal (Table 4). Moreover, the formation rate of those lipid oxidation products apparently
298
decreased in the microcapsules coated by SPI-Md conjugate hydrolysates during the storage. The
299
hydroperoxide and propanal content after four-week storage amounted to 4.72 mmol/kg oil and 2688 in the
300
microcapsules prepared by SPI-Md conjugate hydrolysates (DH 2.9%), while the corresponding products in
301
the sample encapsulated with the SPI-Md conjugates amounted to 24.06 mmol/kg oil and 10496.89,
302
respectively. After eight-weeks storage, the oxidative stability of fish oil powders coated by the SPI-Md
303
conjugate hydrolysate were found also better than those prepared with SPI-Md conjugates or the SPI/Md
Ac ce p
te
d
291
15
Page 15 of 33
304
mixture. Therefore, the combined modified wall material after limited hydrolysis had favorable protective
305
action for the core material. Factors such as the interfacial film thickness and robustness, the emulsion parameters and their ability to
307
scavenge oxygen and free radicals, were all reported to play roles in oxidative stability [8]. However, in our
308
study, the degrees of hydrolysis (DH) may not be enough to produce sufficient redox compounds for the
309
antioxidant action. We supposed that the main antioxidant effect was due to the oxygen barrier effect of the
310
combined modified SPI products by forming the strong surface films surrounding the oil droplet, as well as
311
the tightly packed and arranged hydrogen-bonded network structure of the polysaccharide component,
312
which was consistent with the conclusion of Atarés et al [39, 40]. Especially, the improved emulsifying
313
property of the SPI-Md conjugate hydrolysates after limited hydrolysis was conducive to the
314
microencapsulation through decreasing the oil droplet size, preventing the coalescence during the drying
315
process and making microcapsules more complete. Many authors confirmed that retention of active
316
compounds during the drying process could be enhanced by reducing the emulsion Z-average diameter of
317
dispersed components during emulsification [6, 41, 42]. Our work has demonstrated that combined modified
318
SPI products can be used as effective emulsifying and encapsulating materials for the protection of the
319
oxidizable fish oil.
320
3.2.3 Thermal analysis
Ac ce p
te
d
M
an
us
cr
ip t
306
321
Thermogravimetric analysis (TG) and Derivativ Thermogravimetry (DTG) graphs for microparticles
322
encapsulated by the SPI-Md conjugate hydrolysate (DH 2.9%), SPI-Md conjugate, the mixture of SPI and
323
Md, as well as the fish oil are shown in Fig. 3. No degradation of fish oil was observed below 150 oC.
324
However, almost 100% mass loss for fish oil with complete decomposition occurred in the range between
325
200 and 250 °C, and with maximal weight loss at 210 °C as shown in Fig. 3D-F. From the TG and DTG
326
curves of the fish oil microparticles stabilized by the SPI-Md conjugate hydrolysate (DH 2.9%)，slight mass 16
Page 16 of 33
loss related to residual water molecules was observed below 200 oC. The visible decrease of the TG curves
328
occurred at 250 to 500 oC, which was attributed to the degradation of modified SPI products and some core
329
materials. The degradation of the wall material has been found with whole mass loss from 200 to 500 oC
330
(Fig. 3D), corresponding to the thermal behavior described by other authors for soy protein [6, 43-45]. In the
331
temperature range of 650 to 900 oC, further degradation of G-H 2.9% microparticles was still observed.
332
Especially, more than 18% mass loss after 600 oC was detected for the G-H 2.9% microparticles, indicating
333
that the microcapsules possessed strong heat resistance. However, the microcapsules coated by the SPI-Md
334
conjugates presented different mass loss behavior as the first degradation took place at a temperature range
335
of 100-210 oC (Fig. 3B and E). This was mainly related to the loss of the fish oil on the surface or from the
336
cracks of the microcapsules, since no degradation of the wall material was observed. The main mass loss
337
was found at the temperature range of 210-500 oC, which was attributable not only to fish oil thermal
338
degradation, but also to the initial degradation of SPI-Md conjugate. The microcapsules using the SPI/Md
339
mixture as wall material formed a broad peak ranging from 100 to 500 oC (Fig. 3C and F). Overall, the
340
microcapsule prepared by the SPI-Md conjugates and SPI/Md mixture both showed the core material
341
degradation below 200 oC, indicating poor thermal stability and inefficient microencapsulation, which was
342
consistent with the results of microencapsulation efficiency (Table 4). Nevertheless, the advantage of the
343
microcapsule of SPI-Md conjugate hydrolysate (DH 2.9%) in the thermal stability was apparently due to the
344
increased emulsifying property after limited hydrolysis, thus completely coating the fish oil droplets.
345
3.2.4 Scanning Electron Microscope (SEM) micrograph
Ac ce p
te
d
M
an
us
cr
ip t
327
346
Freeze drying was used to prepare the fish oil microcapsules. First, the emulsions were frozen, followed
347
by the removal of water by sublimation. The SEM micrographs (Fig. 4A–F) highlighted the effects of
348
limited hydrolysis of the wall material on the surface morphology of the particles. After freeze-drying
349
images like wood chips or flakes were observed, confirming the morphology previously described by other 17
Page 17 of 33
authors [46, 47]. For the fish oil droplets microencapsulated by the hydrolyzed SPI-Md conjugates (DH,
351
1.8%-4.6%) (Fig. 4A-C), a relatively smooth surface as well as a rather porous and uniform structure was
352
observed. Whereas, the extensive hydrolysis (DH, 5.7%-6.2%) (Fig. 4D-E) resulted in the particles with a
353
rougher surface, accompanied by irregular agglomerates. Moreover, similar rough morphology was
354
observed for the control sample (fish oil particles prepared with SPI-Md conjugates) as shown in Fig. 4F.
355
The porous surfaces observed were possibly formed by cavities left from ice crystals or air bubbles retained
356
during the freezing [46]. This phenomenon can be explained by the fact that better emulsification of the
357
modified SPI products after limited hydrolysis led to the stable emulsions with small oil droplets uniformly
358
coated by the wall material in the water-phase, which was just the resource of the ice crystals, and thus
359
porous and uniform structure formed after the freeze-drying process.
360
Conclusions
d
M
an
us
cr
ip t
350
This study is related to the use of hydrolyzed protein-polysaccharide conjugates in microencapsulation of
362
fish oil by freeze-drying method. Structural modifications of the native SPI were carried out by limited
363
hydrolysis following the glycosylation with maltodextrin to improve its functional properties for the
364
application as an encapsulating agent. The surface and interfacial tension decreased after limited hydrolysis
365
indicating a better amphiphilic property, which was further confirmed by the decrease of emulsion droplet
366
size as well as the PDI value. Emulsions showed shear-thinning behavior of the emulsions, and no creaming
367
or phase separation was observed after three-week storage. Additionally, the storage stability experiments
368
showed that, the oxidative stabilities (POV and headspace propanal) of microcapsule particles prepared by
369
the modified SPI products with limited hydrolysis were apparently improved. However, extensive hydrolysis
370
might generate shorter proteinic chain length, concomitantly, decreasing the MEE. The knowledge reported
371
in this study would be useful in the development of new delivery systems for bioactive substances.
Ac ce p
te
361
18
Page 18 of 33
372
373
Acknowledgments
This research was financially supported by projects of the National 125 Program of China (2011BAD23B04, 2012BAD33B05 and 2013AA102204).
375
References
376
[1] Y. Kagami, S. Sugimura, N. Fujishima, K. Matsuda, T. Kometani and
377
(2003) 2248.
378
[2] M. Oarada,
379
[3] K. Heinzelmann, K. Franke, J. Velasco and G. Márquez-Ruiz, Eur Food Res Technol, 211 (2000) 234.
380
[4] M.-Y. Baik, E.L. Suhendro, W.W. Nawar, D.J. McClements, E.A. Decker and
381
Chem Soc, 81 (2004) 355.
382
[5] S. Drusch, Y. Serfert, A. Van Den Heuvel and
383
[6] A. Nesterenko, I. Alric, F. Silvestre and
384
[7] M. Akhtar and
385
[8] J.K. Rusli, L. Sanguansri and
386
[9] L.S.R. Sanguansri, Werribee, VIC 3030, AU) and
387
AU) in, Commonwealth Scientific and Industrial Research Organisation (Limestone Avenue, Campbell,
388
ACT 2612, AU) 2012.
389
[10] M.A. Augustin, L. Sanguansri and
390
[11] C. Chung, L. Sanguansri and
391
[12] Y. Kagami, S. Sugimura, N. Fujishima, K. Matsuda, T. Kometani and Y. Matsumura2, J Food Sci,
392
68 (2003) 2248.
cr
ip t
374
P. Chinachoti, J Am Oil
M
an
J Am Oil Chem Soc, 39 (1990) 373.
d
K. Schwarz,
te
V. Durrieu,
Food Res Int, 39 (2006) 807.
Food Res Int, 48 (2012) 387.
"Colloids Surf, B ", 31 (2003) 125.
Ac ce p
E. Dickinson,
J Food Sci, 68
us
Y. Matsumura,
M.A. Augustin,
J Am Oil Chem Soc, 83 (2006) 965. M.A.C.C. Augustin, Wheelers Hill, Victoria 3150,
O. Bode2,
J Food Sci, 71 (2006) E25.
M.A. Augustin,
Food Biophys 3(2008) 140.
19
Page 19 of 33
393
[13] M. Akhtar and
E. Dickinson,
Food Hydrocoll, 21 (2007) 607.
394
[14] S. Drusch, S. Berg, M. Scampicchio, Y. Serfert, V. Somoza, S. Mannino and
395
Hydrocoll, 23 (2009) 942.
396
[15] R. Amarowicz,
397
[16] A. Karayannidou, E. Makri, E. Papalamprou, G. Doxastakis, I. Vaintraub, N. Lapteva and
398
Food Chem, 104 (2007) 1728.
399
[17] X. Guan, H. Yao, Z. Chen, L. Shan and
400
[18] Q. Zhao, H. Xiong, C. Selomulya, X.D. Chen, H. Zhong, S. Wang, W. Sun and
401
Chem, 134 (2012) 1360.
402
[19] J. O'Regan and
403
[20] J.B. Zhang, N.N. Wu, X.Q. Yang, X.T. He and
404
[21] S. Petruccelli and
405
[22] H. Frister, H. Meisel and
406
631.
407
[23] J. Adler-Nissen, Elsevier Applied Science Publishers, 1986.
408
[24] R. Liang, S. Xu, C.F. Shoemaker, Y. Li, F. Zhong and
409
7548.
410
[25] A. Millqvist-Fureby,
411
[26] U. Klinkesorn, P. Sophanodora, P. Chinachoti, E.A. Decker and
412
(2006) 449.
413
[27] H. W, in:
414
ed, AOAC International Gaithersberg, MD, 2002.
415
[28] A. Nesterenko, I. Alric, F. Violleau, F. Silvestre and
K. Schwarz, Food
Eur J Lipid Sci Tech, 112 (2010) 695.
Food Chem, 101 (2007) 163.
M.C. Anon,
Food Chem, 123 (2010) 21. L.J. Wang,
Food Hydrocoll, 28 (2012) 301.
M
D.M. Mulvihill,
Q. Zhou, Food
an
us
M. Zhang,
cr
ip t
G. Articov,
J Agric Food Chem, 43 (1995) 2001.
Ac ce p
te
d
E. Schlimme, Fresenius Zeitschrift fur Analytische Chemie, 330 (1988)
Q. Huang, J Agric Food Chem, 60 (2012)
"Colloids Surf, B ", 31 (2003) 65. D.J. McClements,
Food Res Int, 39
Peroxide value of oils and fats, Official methods of analysis of AOAC international, 17th
V. Durrieu,
Food Res Int, 53 (2013) 115.
20
Page 20 of 33
416
[29] Y. Zhang, C. Tan, X. Zhang, S. Xia, C. Jia, K. Eric, S. Abbas, B. Feng and
417
Technol, (2014) 10.1007/s00217.
418
[30] L. Chen, J. Chen, J. Ren and
419
[31] M.P. Rascón, C.I. Beristain, H.S. García and
420
[32] K.D. Martínez, C. Carrera Sánchez, J.M. Rodríguez Patino and
421
(2009) 2149.
422
[33] M. Corzo-Martínez, C. Carrera Sánchez, F.J. Moreno, J.M. Rodríguez Patino and
423
Food Hydrocoll, 27 (2012) 438.
424
[34] M. Corzo-Martínez, C. Carrera-Sánchez, M. Villamiel, J.M. Rodríguez-Patino and
425
Food Eng, 113 (2012) 461.
426
[35] J. Maldonado-Valderrama and
J.M.R. Patino,
427
[36] Y. Wang, D. Li, L.-J. Wang and
B. Adhikari,
428
[37] R. Pal,
429
[38] P. Fäldt and
430
[39] L. Atarés, J. Bonilla and
431
[40] A.I. Bourbon, A.C. Pinheiro, M.A. Cerqueira, C.M.R. Rocha, M.C. Avides, M.A.C. Quintas and
432
Vicente,
433
[41] A. Soottitantawat, H. Yoshii, T. Furuta, M. Ohkawara and
434
[42] N. Mongenot, S. Charrier and
435
[43] H. Wang, L. Jiang and
436
[44] V. Schmidt, C. Giacomelli and
437
[45] P. Guerrero, A. Retegi, N. Gabilondo and
438
[46] M.C. Farias, M.L. Moura, L. Andrade and
Eur Food Res
Food Hydrocoll, 25 (2011) 887. Food Sci Technol Res , 44 (2011) 549. A.M.R. Pilosof, Food Hydrocoll, 23
ip t
M.A. Salgado,
M. Villamiel,
F.J. Moreno,
J
an
us
cr
M. Zhao,
F. Zhong,
M
Curr Opin Colloid Interface Sci, 15 (2010) 271. J Food Eng, 104 (2011) 56.
Food Sci Technol Res, 29 (1996) 438.
te
B. Bergenståhl,
d
Curr Opin Colloid Interface Sci, 16 (2011) 41.
J Food Eng, 100 (2010) 678.
Ac ce p
A. Chiralt,
A.A.
J Food Eng, 106 (2011) 111.
L. Fu,
a.P. Chalier,
P. Linko,
J Food Sci, 68 (2003) 2256.
J Agric Food Chem, 48 (2000) 861.
J Appl Polym Sci, 106 (2007) 3716. V. Soldi,
Polym Degrad Stab, 87 (2005) 25. K. de la Caba,
J Food Eng, 100 (2010) 145.
M.H.M.R. Leão,
Mater Res, 10 (2007) 57.
21
Page 21 of 33
439
[47] M. Sousdaleff, M.L. Baesso, A.M. Neto, A.C. Nogueira, V.A. Marcolino and
440
Food Chem, 61 (2013) 955.
G. Matioli, J Agric
441 442
Ac ce p
te
d
M
an
us
cr
ip t
443
22
Page 22 of 33
443
Figure captions
444
Fig. 1 Surface (A) and interfacial (B) tension of combined modified SPI products and native SPI as a
446
function of concentration. The points in the graphic are means ± standard deviation of three separate
447
determinations (mean± SD, n = 3).
448
Fig. 2 Effects of different degree of hydrolysis (DH) of combined modified SPI products on the rheological
449
properties of emulsions with fish oil/protein ratio of 0.67.
450
Fig. 3 The thermogravimetric weight loss (TG) (A, B, C) and derivative of weight loss (DTG) (D, E, F)
451
curve of the fish oil microparticles stabilized by the modified SPI products.
452
Fig. 4 Scanning electron micrographs of freeze dried fish oil microparticles stabilized by various combined
453
modified SPI products (A-E: DH 1.8%, DH 2.9%, DH 4.6%, DH 5.7%, DH 6.2%; E: SPI-Md).
Ac ce p
te
d
M
an
us
cr
ip t
445
23
Page 23 of 33
*Graphical Abstract (for review)
Ac
ce pt
ed
M
an
us
cr
ip t
Graphical Abstract
Page 24 of 33
*Highlights (for review)
Highlights Wall material of microcapsules is crucial for protecting fish oil from oxidation.
Hydrolysate of SPI-Maltodextrin conjugate gives better stabilization.
Degree of hydrolysis strongly affects the emulsions and powders of fish oil.
Emulsions coated by limited hydrolyzed products showed better emulsion stability.
Powders coated by hydrolysates (DH 2.9%) showed better oxidative stability.
Ac
ce pt
ed
M
an
us
cr
ip t
Page 25 of 33
Table(s)
Tables
Table 1 The total solids, viscosity, particle size and size distribution of combined modification SPI/fish oil emulsion after high pressure homogenizationa
7.51±0.06 8.00±0.26a 8.30±0.17a 10.90±0.27b 12.20±0.55c
a
3.69±0.05 3.75±0.06a 4.63±0.07c 4.09±0.17b 4.91±0.05c
Z-average (nm) b
273.0±2.52 317.7±4.39c 235.5±0.29a 282.1±1.16b 407.4±5.20d
PDI 0.315±0.002c 0.204±0.010b 0.162±0.001a 0.203±0.005b 0.374±0.007d
The emulsions stabilized by combined modified SPI products (DH 5.7%), with two-stage homogenization (35+10
us
a
a
Viscosity (mPas)
ip t
0.25 0.43 0.67 1.0 1.5
Total solids (% w/w)
cr
Oil/Protein
Ac
ce pt
ed
M
an
MPa). All emulsions were prepared in triplicate. Different lower case letters are significantly different (P < 0.05).
Page 26 of 33
Table 2 The total solids, viscosity, particle size and size distribution of combined modified SPI/fish oil emulsiona initially prepared and 3 weeks (4 oC) after high pressure homogenization
a
9.90±0.12c 11.4±0.58d 8.90±0.26ab 8.30±0.23a 9.70±0.12bc 8.55±0.29a 9.81±0.12bc
4.05±0.06a 5.32±0.08c 4.00±0.29a 4.63±0.18b 5.10±0.17c 3.56±0.12a 4.61±0.12b
Initially prepared Z-average (nm) PDI a 212.5±1.01 0.115±0.001ab 232.6±1.85b 0.091±0.021a d 329.3±2.37 0.193±0.023c 253.6±2.92c 0.154±0.011b 254.8±0.27c 0.116±0.004ab 401.3±4.61f 0.293±0.010d 344.5±4.32e 0.279±0.004d
After 3 weeks storage Z-average (nm) PDI a 251.3±1.12 0.166±0.005a 255.6±1.15a 0.154±0.006a b 266.6±1.15 0.167±0.004a 281.5±0.58c 0.157±0.004a 288.5±1.01c 0.156±0.006a 585.2±5.85e 0.490±0.011b 491.6±5.60d 0.419±0.003c
ip t
Viscosity (mPas)
cr
Sample 1.80% 2.90% 4.60% 5.70% 6.20% SPI/Md(2/1)a SPI-Md(2/1)b
Total solids (% w/w)
The emulsions were prepared with the oil/protein ratio of 0.67/1 and two-stage homogenization (35+10 MPa). All
SPI-Md (2/1) was the conjugates of the SPI and Maltodextrin at 80 oC for 80 min with the ratio of 2/1. Different
an
c
us
emulsions were prepared in duplicate. b SPI/Md (2/1) was the mixture of SPI and Maltodextrin with the ratio of 2/1.
Ac
ce pt
ed
M
lower case letters are significantly different (P < 0.05).
Page 27 of 33
Table 3 Rheological parameters obtained from fitting using Power Law Model for fish oil emulsions prepared with different modified SPI products K (Pa·sn)
n
R2
1.8% 2.9% 4.6% 5.7% 6.2%
0.011±0.001 0.014±0.001 0.010±0.001 0.013±0.001 0.017±0.001
0.831±0.014 0.825±0.010 0.833±0.014 0.810±0.011 0.773±0.007
0.997 0.998 0.996 0.998 0.999
Ac
ce pt
ed
M
an
us
cr
ip t
Sample
Page 28 of 33
Table 4 The MEE and oxidative stability (headspace propanal and POV)a of the combined modification SPI/fish oil microcapsules Initially prepared Sample
MEE (%)
4 Weeks
Headspace POV
propanal
(mmol/kg)
8 Weeks
Headspace Retention
POV
time(min)
(mmol/kg)
(GC area)
propanal
Headspace Retention
POV
time(min)
(mmol/kg)
(GC area)
Retention time(min)
(GC area)
97.84
1.10
1050.98
1.057
9.77
5750.49
1.058
12.60
8728.72
1.034
2.90%
95.08
1.77
537.60
1.055
4.72
2688.00
1.054
8.06
2980.59
1.040
4.60%
82.22
1.66
1301.84
1.060
12.90
5069.18
1.060
5.70%
90.52
1.99
1732.73
1.058
17.45
6863.65
1.057
91.47
1.99
1421.02
1.061
24.54
7105.09
1.061
66.31
2.42
2861.60
1.052
24.06
10848.27
1.053
SPI-Md (2/1)
26.93
2.36
2909.38
1.054
24.06
10496.89
1.054
16.67
9646.57
1.042
19.32
9191.57
1.037
27.79
9040.13
1.042
27.82
16691.58
1.041
28.11
20306.12
1.040
cr
6.20%
ip t
1.80%
SPI/Md (2/1)
ce pt
ed
M
an
us
All the values shown in the table were the means of three parallel experiments (n=3).
Ac
a
propanal
Page 29 of 33
Figure(s)
Figure
Ac
ce pt
ed
M
an
us
cr
ip t
Fig. 1
Page 30 of 33
Ac
ce pt
ed
M
an
us
cr
ip t
Fig. 2
Page 31 of 33
Ac
ce pt
ed
M
an
us
cr
ip t
Fig. 3
Page 32 of 33
Ac
ce pt
ed
M
an
us
cr
ip t
Fig. 4
Page 33 of 33