Accepted Manuscript Grape seed and apple tannins: emulsifying and antioxidant properties Maria Cruz Figueroa-Espinoza, Andrea Zafimahova, Pedro G. Maldonado Alvarado, Eric Dubreucq, Céline Poncet-Legrand PII: DOI: Reference:
S0308-8146(15)00058-8 http://dx.doi.org/10.1016/j.foodchem.2015.01.056 FOCH 17007
To appear in:
Food Chemistry
Received Date: Revised Date: Accepted Date:
30 September 2014 7 January 2015 8 January 2015
Please cite this article as: Figueroa-Espinoza, M.C., Zafimahova, A., Maldonado Alvarado, P.G., Dubreucq, E., Poncet-Legrand, C., Grape seed and apple tannins: emulsifying and antioxidant properties, Food Chemistry (2015), doi: http://dx.doi.org/10.1016/j.foodchem.2015.01.056
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
Grape seed and apple tannins: emulsifying and antioxidant properties †
2
Maria Cruz Figueroa-Espinoza1, Andrea Zafimahova1,2,3,4, Pedro G. Maldonado Alvarado1,2,3,4,
3
Eric Dubreucq1, Céline Poncet-Legrand2,3,4,*
4 5 6 1
7 8
2
Montpellier SupAgro, UMR1208 IATE, F-34060 Montpellier, France
INRA, 3 Montpellier SupAgro, 4 Université Montpellier II, UMR1083 SPO, F-34060 Montpellier,
9
France
10 11 12 13
*
Corresponding author: [email protected]
14 15 16 17
† Dedicated to the memory of Gérard Mazerolles
18
1
19
Abstract
20 21
Tannins are natural antioxidants found in plant-based foods and beverages, whose amphiphilic
22
nature could be useful to both stabilize emulsions and protect unsaturated lipids from oxidation. In
23
this paper, the use of tannins as antioxidant emulsifiers was studied. The main parameters
24
influencing the stability of emulsions (i.e. tannins structure and concentration, aqueous phase pH,
25
and ionic strength) were identified and optimized. Oil in water emulsions stabilized with tannins
26
were compared with those stabilized with two commercial emulsifying agents, poly(vinyl alcohol)
27
(PVA) and polyoxyethylene hydrogenated castor oil. In optimized conditions, the condensed
28
tannins allowed to obtain a stability equivalent to that of PVA. Tannins presented good antioxidant
29
activity in oil in water emulsion, as measured by the Conjugated Autoxidizable Triene (CAT)
30
assay.
31 32
Keywords: antioxidant, polyphenols, tannins, catechin, emulsion, CAT assay, poly(vinyl alcohol).
33 34
2
35
1. Introduction
36
Emulsions are formed when two immiscible liquids are mixed. One liquid (the dispersed
37
phase) is dispersed as droplets (from 0.10 µm to a few µ m of diameter) in the other (the continuous
38
phase) (McClements, 2007). Forming emulsions requires energy to increase the interfacial area
39
between continuous and dispersed phases. Emulsified systems are thermodynamically unstable
40
because of the surface tension between oil and water, which opposes to the increase of interfacial
41
area. Emulsions can be stabilized by amphiphilic molecules, which adsorb in the oil-water
42
interface, decreasing the surface tension. They can also be stabilized by solid particles which adsorb
43
onto the interface between the two phases and are called Pickering emulsions (Pickering, 1907).
44
Besides, the stability of emulsions is conditioned by the competition between attractive (Van der
45
Waals, hydrophobic interactions, electrostatic attractions, hydrogen bonds) and repulsive forces
46
(electrostatic repulsion, steric repulsion) between the dispersed droplets (Guzey & McClements,
47
2007; Tcholakova, Denkov, Sidzhakova, Ivanov, & Campbell, 2005). Stabilizing agents increase
48
repulsions between droplets and slow down coalescence and phase separation phenomena. The
49
stabilization of emulsions also depends on constituents of the emulsion, namely the concentration in
50
emulsifier or stabilizing agent, pH, viscosity and ionic strength of the aqueous phase, and the
51
concentration of the organic phase (Chanamai & McClements, 2000; Tcholakova, Denkov,
52
Sidzhakova, & Campbell, 2006). Emulsified systems raise the problem of oxidation: dispersal of
53
lipids in emulsified systems increases the specific area in contact with oxygen and some pro-
54
oxidizing species (Coupland & McClements, 1996), which are detrimental to lipids.
55
On another hand, polyphenols are known for their antioxidant properties. They are molecules found
56
in large amounts in plant-based foods and beverages. They are constituted by one or several rather
57
hydrophobic aromatic nuclei bearing polar hydroxyl groups (-OH) (Poncet-Legrand, Cartalade,
58
Putaux, Cheynier, & Vernhet, 2003). Among them, a particular class of polyphenols called tannins
59
is produced in large amounts by distilleries and wine industries. These compounds are polymers of
3
60
flavan-3-ol units primarily linked by C4-C8 bonds, with C4-C6 bonds giving rise to some degree of
61
polymer branching (Figure 1). For example, the constitutive units of grape seed tannins are (+)-
62
catechin (C), (-)-epicatechin (Ec), and (-)-epicatechin gallate (EcG) (Prieur, Rigaud, Cheynier, &
63
Moutounet, 1994). Conversely, apple condensed tannins present an homogeneous structure, with
64
primarily (-)-epicatechin as the constitutive flavanol unit (Guyot, Doco, Souquet, Moutounet, &
65
Drilleau, 1997). The antibacterial tannin activity is related to surface chemistry and to their ability
66
to non-covalently associate with proteins and other macromolecular structures (Vidal, Francis,
67
Guyot, Marnet, Kwiatkowski, Gawel, et al., 2003).
68
Their structure (aromatic hydrophobic rings, hydroxyl hydrophilic groups) suggests that some
69
tannins may have surface active properties, and might thus stabilize emulsions. Although there are
70
already numerous natural surfactant agents, the interest to use tannins as stabilizers lies in the fact
71
that they also have important antioxidant capacities. They are highly polymerized and possess many
72
phenolic hydroxyl groups. In addition, the “B” ring in the flavanols is responsible for most of the
73
antioxidant activity, as it contains the catechol or trihydroxy functionality. (Hagerman, Riedl,
74
Jones, Sovik, Ritchard, Hartzfeld, et al., 1998; Pazos, Gallardo, Torres, & Medina, 2005; Torres,
75
Centelles, Cascante, Xavier, & Bobet, 2002; Van Acker, Van Den Berg, Tromp, Griffioen, Van
76
Bennekom, Van Der Vijgh, et al., 1996). However, due to their high chemical reactivity, they
77
undergo numerous reactions when they are in solution, leading to the formation of new compounds
78
which possess chemical structures, conformation in solution, and water solubility different from
79
those of native tannins.
80
In this paper, the following properties of catechin, grape seed and apple tannins were studied: oil in
81
water emulsion stabilization capacity and antioxidant activity in emulsified system. The effect of
82
the tannin’s structure (chemical composition and polymerization degree) and concentration, as well
83
as of pH and ionic strength on the emulsion stability was also investigated.
84
4
85
2. Materials and Methods
86
2.1.
Materials
87
Methyl oleate (the oil phase) and Eumulgin ® HRE40, an ether of polyoxyethylene
88
cetyl stearyl alcohol, were supplied by Cognis (Monheim, Germany; now BASF). Phosphate
89
buffers (pH 3 to 9, ionic strengths 50 to 100 mM) were prepared in deionised water (MilliQ system,
90
Millipore, USA) from reagents obtained from Sigma Aldrich (St. Louis, MO, USA). Poly(vinyl
91
alcohol) (PVA) was purchased from Merck (Whitehouse Station, NJ, USA).
92
In this paper, tannins will be referred to as Gn or An (repectively for Grape and Apple),
93
where n stands for the number average degree of polymerization, followed by the suffix “ox” when
94
they were oxidized. The polyphenols used in this study were a commercial monomer (catechin,
95
DP1) supplied by Sigma (St. Louis, MO), tannin fractions purified from Grape seed (Vitis vinifera,
96
var. Shiraz) (G4ox and G15ox) obtained as referred in the literature (Poncet-Legrand, Cartalade,
97
Putaux, Cheynier, & Vernhet, 2003), and Apple (Malus sylvestris var. Kermerrien) (A6 and 15) as
98
in reference (Michodjehoun-Mestres, Souquet, Fulcrand, Meudec, Reynes, & Brillouet, 2009).
99
Their degree of polymerization was determined by depolymerisation followed by HPLC analysis,
100
as described by Preys et al. (Preys, Mazerolles, Courcoux, Samson, Fischer, Hanafi, et al., 2006)
101
immediately after their purification. Depolymerisation was also performed prior to emulsion
102
preparation, and we observed a drop in depolymerisation yield compared to the initial one with the
103
grape seed fractions (but not the apple tannins) and this was attributed to oxidation. In this case, the
104
characterization of fractions by standards methods becomes inaccurate (Mouls & Fulcrand, 2012;
105
Poncet-Legrand, Cabane, Bautista-Ortín, Carrillo, Fulcrand, Pérez, et al., 2010; Vernhet,
106
Dubascoux, Cabane, Fulcrand, Dubreucq, & Poncet-Legrand, 2011): the effective degrees of
107
polymerization of these oxidized tannins is larger. In a second set of experiments, emulsions were
108
prepared with freshly purified apple tannins (A15b). Oxidized catechin (D1ox) was obtained by
109
stocking a catechin (D1) solution (3 mg mL-1 in phosphate buffer pH 6.5, 50 mM) at 28°C for 7
110
days. To obtain oxidized apple tannins (A15box), 33 mg of apple tannins (A15b) were solubilized 5
111
in 11 mL of filtered (0.22 µm) phosphate buffer (pH 6.5, 50 mM), and left in contact with air
112
during 15 days at 28 °C. The buffers were filtered in order to minimize the risks of bacterial
113
growth. Oxidized tannins exhibit a different color from the non-oxidized ones: their color in
114
solution is ranging from dark yellow to brown. Depending on the application (food or non-food
115
application) these color changes may or may not be accepted from a consumer point of view.
116
2.2.
117
Methods 2.2.1. Emulsion preparation
118
Tannins were dissolved in the adequate phosphate buffer at concentrations ranging from
119
0.1 to 1 % (w/w), 24 hours prior to emulsion preparation. Methyl oleate (10 % v/v of the total
120
emulsion volume) was then added to the aqueous phase. Preparation of the emulsion was performed
121
by sonicating the mixture during 30 s at room temperature (20°C) using an output signal amplitude
122
of 15 µm (peak to peak), placing the probe sonicator (Sonifier 250, Branson Ultrasonics Co., Paris,
123
France) vertically in the centre of the recipient, about 5 mm from the bottom.
124
2.2.2. Emulsion properties
125
Conductivity measurements were done using a conductimeter to determine whether the
126
emulsions were direct (the continuous phase is water; high conductivity) or inverse (the continuous
127
phase is oil; zero conductivity). The granulometric analysis of emulsions was done with a light
128
scattering instrument (Mastersizer 2000 laser granulometer, Malvern, Malvern, UK). Stability was
129
monitored over time by measuring the emulsified volume (Ve) compared to the total volume (Vt).
130
2.2.3. Antioxidant properties of tannins
131
Antioxidant capacity of tannin fractions was measured using the CAT (conjugated
132
autoxidizable triene) procedure developed by Laguerre et al. (Laguerre, Lopez-Giraldo, Lecomte,
133
Barea, Cambon, Tchobo, et al., 2008). Briefly, this assay is based upon the high sensitivity to
134
oxidation of the α-eleostearic acid present in the triacylglycerols of Tung (Vernicia fordii) oil.
6
135
Eleostearic acid (ELA) is a linolenic acid containing a conjugated triene part: it thus has a strong
136
UV absorbance at λmax = 273 nm. Upon oxidation, the degradation of the conjugated triene system
137
into a conjugated diene is accompanied by a decrease in the signal at 273 nm (and an increase at
138
234 nm). Tung oil oxidation was initiated by 2,2’-azobis(2-amidinopropane) dihydrochloride
139
(AAPH) in a water in oil emulsion at 37 °C, in the presence or absence of the tested antioxidant.
140
Spectrophotometric monitoring of the UV signal decay at 273 nm determines the ability of a
141
molecule to protect tung oil from oxidation. Trolox (6-hydroxy-2,5,7,8-tetramethylchroman-2-
142
carboxylic acid) was used as reference antioxidant. The CAT value is expressed as micromoles of
143
trolox equivalents (TE) per gram of dried tannins.
144
2.2.4. Experimental design to optimize emulsion stability
145
Emulsion stability was monitored overtime by measuring the emulsified volume (Ve) with regards
146
to the total volume (Vt) (McClements, 2007; Novales, Papineau, Sire, & Axelos, 2003). A Doehlert
147
experimental design (Doehlert, 1970) was built for determining the experimental conditions that
148
provide the highest emulsion volume using three factors (pH 4.0-9.0; tannin concentration (0.1-0.5
149
% w/w), ionic strength (50-100 mM)) at 7, 6, and 5 levels respectively. A total of 13 experiments
150
were realized, including three replicates at the central point (Table 1). The total amount of
151
emulsion was 10 mL, prepared with 10 % of oil.
152
The model to which the experimental data were fitted was a second-order polynomial
153
model. The following equation was used:
154
Y = a0 + a1 X1 + a2 X2 + a3 X3 + a12 X1X2 + a13 X1X3 + a23 X2X3 + a11 X12 + a22 X22 + a33 X32
155
where Y is the emulsified volume, ai are the regression coefficients, and X1, X2, and X3 are the
156
experimental factors. Confidence intervals were evaluated with T95%,3 as the Student coefficient.
157
7
158
3. Results and discussion
159
Tannins present a huge diversity of structures (monomer composition, degree of polymerization
160
(DP), branching). Prior to the optimization of pH and ionic strength, two important parameters in
161
emulsion stabilization (Chanamai & McClements, 2000; Gu, Decker, & McClements 2004), a
162
preliminary study of the effect of the polymer structure on emulsion stability was performed. The
163
emulsifying properties of catechin (DP1) and oxidized catechin (DP1ox), two fractions of grape
164
seeds tannins (DP4 and 15, named G4ox and G15ox) and of two fractions of apple tannins (DP6
165
and 15, named A6 and A15), were compared. Indeed apple and grape seed tannins differ in
166
composition: apple tannins are homopolymers of epicatechin, while grape seed tannins are random
167
copolymers of catechin, epicatechin and epicatechin gallate (Figure 1).
168
After determining which tannins were the best stabilizers, tannin concentration, pH, and
169
ionic strength, were optimized and a comparison was made with two commercial emulsifiers (PVA
170
and Eumulgin HRE40).
171 172
3.1.
173
degree)
Effect of tannin structure (chemical composition, degree of polymerization, oxidation
174
The effect of tannin structure was studied by preparing emulsions with catechin
175
(monomer), apple and grape seed tannin fractions having a different DP’s. Oil in water emulsions
176
were prepared in phosphate buffer (50 mM, pH 7.0) with 10 % of methyl oleate, and emulsifier
177
(0.1 % w/v of the water phase). After emulsification by sonication, the turbidity of the emulsion in
178
the transparent tube was the same on all the height of the tube. Emulsion destabilization was then
179
monitored during at least 3 days at room temperature and the ratio emulsified volume/total volume
180
was plotted as a function of time (Figure 2).
181
A phase separation occurred very quickly (within two minutes) for reference emulsions (i.e. methyl
182
oleate and buffer without added emulsifier). The same behavior was observed for the emulsions 8
183
stabilized with catechin (DP1) and oxidized catechin (DP1ox), not shown, as well as apple tannins,
184
except for A6, which provided a good stabilisation during the first hour. After 24 hours, emulsified
185
volumes obtained with Eumulgin were superior to those obtained with the PVA (97 % vs 75 %) and
186
grape seed tannins. For these two last ones, emulsified volumes were higher with G15ox (95 %)
187
than with G4ox (30 %) after 24 hours. These values were the same after 3 days, except for G15ox:
188
emulsion represented 80 % of the volume. Grape seeds tannins with the highest DPs (G15ox)
189
provided the best stabilization.
190
Tannins are thus able to stabilize emulsions under certain conditions. In the conditions
191
tested, their chemical structures had a strong impact on their physicochemical properties: grape
192
tannins stabilizing properties increased with their DP, and for a given DP, galloylated tannins (i.e.
193
grape seeds tannins, having ~ 25 % (w/w) of epicatechin gallate units) seemed to be more efficient
194
than apple tannins.
195
However, the grape seed tannin fraction used was oxidized and it was thus difficult to
196
compare apple and grape seed tannins. Upon oxidation, tannins undergo many chemical reactions
197
and some of these reactions lead to the formation of higher molecular weight species (Poncet-
198
Legrand, Cabane, Bautista-Ortín, Carrillo, Fulcrand, Pérez, & Vernhet, 2010; Vernhet, Dubascoux,
199
Cabane, Fulcrand, Dubreucq, & Poncet-Legrand, 2011). Furthermore, the conformation of the
200
macromolecules changes (e.g. from linear to branched macromolecules or from semi-flexible to
201
more rigid chains), and water-solubility of the fractions also changes. Analysis of oxidized tannins
202
(real DP, conformation in solution) is not easy and when analyzing G15ox by Small Angle X-Ray
203
Scattering (Vernhet, Dubascoux, Cabane, Fulcrand, Dubreucq, & Poncet-Legrand, 2011), results
204
showed that the actual degree of polymerization of grape seed tannin was 25 rather than 15. The
205
impact of oxidation was confirmed by preparing emulsions with apple tannins (A15b) just after
206
they were extracted and purified, and after oxidation in phosphate buffer 50 mM, pH 6.5 (A15box):
207
unoxidized tannins did not stabilize emulsions, whereas the oxidized ones did (results not shown).
9
208
Emulsion stabilizers are often surface active agents that are anchored at the interface
209
between oil and aqueous solution to reduce surface tension. It was thus verified if tannins were
210
surface active by measuring surface tension of tannin solution: at the concentration used in this
211
study, the surface tension was not significantly decreased, meaning that in our conditions (ionic
212
strength, pH, concentration), the tannins used in this study were not surface active (Supplementary
213
data S1).
214
The mechanism of emulsion stabilization by tannins may be the result of steric and/or
215
electrostatic repulsions between droplets covered by adsorbed polymers. Last, but not least,
216
emulsions can also be stabilized by particles (Pickering, 1907), and tannins, especially the oxidized
217
ones, have been reported to form aggregates and particles (Poncet-Legrand, Cartalade, Putaux,
218
Cheynier, & Vernhet, 2003; Zanchi, Vernhet, Poncet-Legrand, Cartalade, Tribet, Schweins, et al.,
219
2007). The stabilization increased with the tannin degree of polymerization (which remained low in
220
comparison with other types of biopolymers: all tannins used in this study had molecular weight
0.5). pH was the parameter which had more effect on the
242
emulsified volume, as well as its combination with itself and the ionic strength. It is worth noting
243
that the quadratic form of I has also an effect (p values < 0.05).
244
Prediction profiles of the emulsion stability for pH 4, 6, and 8 are shown on Figure 3. At
245
pH 4 (Figure 3A), the emulsified volume increased very slightly with tannin concentration and
246
ionic strength. It is thus likely that the emulsion stability is governed by the steric repulsion
247
between droplets, due to the multilayer tannin adsorption. However, at pH values ranging from 6 to
248
8 (Figure 3A and B), tannin concentration effect is limited and the emulsified volume decreases
249
when the ionic strength increases. Maximum of emulsified volumes would be obtained for values
250
of pH superior to 8 and ionic strengths lower than 50 mM. In this pH range, tannins are more and
251
more negatively charged: catechin and epicatechin pKas are 8.2 and 9.2, respectively (Slabbert,
252
1977). Droplets are mainly stabilized by electrostatic repulsions. Increasing ionic strength at these
253
pHs induce the screening of the charges and thus a lowering of the repulsions. At lower pH (4 for
254
instance), tannins are not charged, and when ionic strength is increasing, they form particles more
255
easily due to hydrophobic interactions (Poncet-Legrand, Cartalade, Putaux, Cheynier, & Vernhet,
256
2003). In the hypothesis of emulsions stabilized by tannin aggregates (Pickering emulsion), this
257
would explain the improved stabilization observed when the ionic strength is increased.
11
258 259
3.3.
Emulsion ageing
260
The effect of grape seeds tannins G15ox on the emulsion ageing was performed by
261
granulometric analysis by monitoring the emulsified volume decrease as a function of time
262
(Supplementary data S2 and S3). Emulsions contained either 50 % (w/w) or 10 % (w/w) of methyl
263
oleate in phosphate buffer (50 mM, pH 7) and tannins G15 (0.9 % w/v of the oil phase). A “fresh”
264
emulsion (one hour old) and a one month old emulsion were compared. In the first experiment
265
(50% of methyl oleate), 10 % of the formed droplets had a size of the order of 0.7 µm. About 50 %
266
of droplets had a size lower or equal to 1 µm and 10 % had a diameter superior to 2.2 µm. After a
267
month of ageing, 10 % of the formed droplets had a size of the order of 2 µm. About 50 % of
268
droplets had a size lower or equal to 3 µm and 10 % had a diameter superior to 5.6 µm. The
269
droplets size thus roughly tripled during the storage.
270
In the second experiment (10% of methyl oleate), less than 10 % of the formed droplets
271
had a size of the order of 0.8 µm. About 50 % of droplets had a size lower or equal to 1.5 µm and
272
10 % had a diameter superior to 3.5 µm. After a month of ageing, the drops average size was about
273
4.6 µm, with a maximal size around 9.2 µm for 10 % of droplets and a minimal size of 3 µm. The
274
droplets size also tripled during the storage.
275
Such a diameter increase may be due to flocculation or to coalescence. In the first case,
276
it would mean that the emulsion degradation is reversible, and that after a certain excitement the
277
emulsion would take back the initial droplets size distribution. In the second case, the degradation
278
is irreversible and it would be necessary to re-emulsify the mixture. Microscopy observations
279
(supplementary data S4) showed that the size of the primary droplets did not increase, whereas
280
aggregates are getting larger after one month. Flocculation is thus the driving mechanism for
281
emulsion destabilization in our systems.
12
282
3.4.
Antioxidant properties
283
Antioxidant properties of different tannins was evaluated by the use of the CAT assay
284
(Laguerre, et al., 2008) using triacylglycerols of tung oil as an ultraviolet probe in an emulsified
285
medium and results are presented in Figure 4. As an example, the kinetics of bleaching in the
286
absence of oxidized apple tannins A15ox in weight equivalents is illustrated. Trolox and all tested
287
grape seed and apple tannins delayed AAPH induced oxidation of stripped tung oil, without
288
exhibiting a significant lag phase. As explained by Laguerre et al. (Laguerre, et al., 2008) this
289
absence of a lag phase is characteristic of molecules that rather scavenge AAPH-derived peroxyl
290
radicals, instead of directly reducing the lipoperoxyl radical derived from oil. Thus Trolox and
291
tannins seem to behave as retarder antioxidants rather than pure chain-breaking antioxidants.
292
Tannins are fairly good antioxidant: grape seed tannins G15ox presented a CAT value of 7200 ±
293
390 TE/ g, followed by non oxidized apple tannins A6 (5490 ± 320 TE/g) and oxidized apple
294
tannins A15ox (3850 ± 430 TE/g). Our results are in the same order of magnitude with the results
295
of previous reported results of 8200 TE/g of (-)-epicatechin and 2530 TE/g of gallic acid
296
(calculated from reference (Laguerre, et al., 2008)). Considering that the location of phenolics
297
toward the oil droplet in an oil in water emulsion is of prime importance in their antioxidant
298
capacities, such a high CAT value for G15ox tannins reflects their good location toward the
299
oxidation site, in this case the lipidic substrate, according to the polar paradox (Porter, Black, &
300
Drolet, 1989). These results are supported by those obtained in section 3.1 showing that grape seed
301
tannins G15ox were the best emulsion stabilizers. Thus, G15ox capacity of adsorbing into the
302
interface may contribute to increase their concentration at the site where oxidation occurs. Besides
303
this, grape seeds tannins, unlike apple tannins whose aromatic rings are dihydroxylated, contain
304
galloylated units with a trihydroxylation pattern. This different hydroxylation substitution, in
305
addition to their high molecular weight and the proximity of many aromatic rings and hydroxyl
306
groups, contribute to increase their antioxidant capacity. The order of antioxidant activity has been
307
reported as (-)-epicatechin < epicatechin-3-O-gallate < (-)-epigallocatechin (Wright, Johnson, &
13
308
DiLabio, 2001), which support our results, grape seed tannins G15ox containing both, (-)-
309
epicatechin and epicatechin gallate units, and apple tannins (A6 and A15) containing only (-)-
310
epicatechin units.
311
Tannins antioxidant properties come from their hydroxyl groups and more specifically
312
from their catechol moieties (i.e. –OH groups on ortho positions on aromatic rings). Grape seed
313
tannins are galloylated and contain more catechol units than the apple ones, which may explain that
314
they are also more antioxidant. When oxidation takes place, it does not lead to the destruction of
315
catechol moieties (S. Guyot, Vercauteren, & Cheynier, 1996) (at least not systematically),
316
explaining why oxidized apple tannins A15ox still have antioxidant activity. This has practical
317
implications, since tannin fractions obtained from distilleries are rarely unoxidized. It is important
318
that they still have antioxidant properties.
319
320
4. Conclusion
321
In this project we investigated the use of phenolic compounds obtained from grape seed
322
and apple as emulsion and antioxidant stabilizers. This first study showed that oxidized grape seed
323
tannins can be used to stabilize oil in water emulsion. Tannin structure plays an important part on
324
emulsion properties: best candidates are relatively high molecular weight tannins, which underwent
325
oxidation reactions (i.e. their chemical structure, and thus their solution properties are different
326
from initial molecules). Emulsions stabilized with tannins were compared with emulsions obtained
327
in the same conditions using model emulsifiers as PVA and with Eumulgin®. In certain
328
physicochemical conditions, the oxidized tannins allowed to obtain a stability equivalent to that of
329
the PVA. The oil in water formed emulsions were up to 50 % of methyl oleate, with drop mean
330
sizes in the micrometer range. Emulsion stabilization appeared to be due to electrostatic repulsions
331
at high pH’s (~ pKa), and emulsion breaking was mainly due to droplet flocculation, as observed by
332
laser granulometry experiments. At pH A6 > A15ox)
337
measured by the CAT test. This research shows the potential use of the winery and distillery by-
338
products and wastes for the production of high added value extracts rich in oxidized tannins that
339
could be used as antioxidants emulsifiers.
340
341 342
Acknowledgements This work was financed by Montpellier SupAgro. We thank Gérard Mazerolles who
343
helped us in the design of experiment and in the interpretation of its results.
344
References
345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371
Cartalade, D., & Vernhet, A. (2006). Polar interactions in flavan-3-ol adsorption on solid surfaces. Journal of Agricultural and Food Chemistry, 54(8), 3086-3094. Chanamai, R. & McClements, D.J. (2000). Dependance of creaming and rheology of monodisperse oil-in-water emulsions on droplet size and concentration. Colloids and Surfaces, 172, 79-86. Coupland, J. N., & McClements, D. J. (1996). Lipid oxydation in food emulsions. Trends in Food Science and technology, 7, 83-91. Doehlert, D. H. (1970). Uniform Shell Designs. Journal of the Royal Statistical Society. Series C (Applied Statistics, 19(3), 231-239. Gu, Y. S., Decker, E.A., & McClements, D.J. (2004). Influence of pH and carrageenan type on properties of β-lactoglobulin stabilized oil-in water emulsions. Food Hydrocolloids, 19, 8391. Guyot, S., Doco, T., Souquet, J.-M., Moutounet, M., & Drilleau, J.-F. (1997). Characterization of highly polymerized procyanidins in cider apple (Malus sylvestris var. kermerrien) skin and pulp. Phytochemistry, 44(2), 351-357. Guyot, S., Vercauteren, J., & Cheynier, V. (1996). Colourless and yellow dimers resulting from (+)catechin oxidative coupling catalysed by grape polyphenoloxidase. Phytochemistry, 42, 1279-1288. Guzey, D., & McClements, D. J. (2007). Impact of electrostatic interactions on formation and stability of emulsions containing oil droplets coated by β-lactoglobulin-Pectin complex. Journal of Agricultural and Food Chemistry, 55, 475-485. Hagerman, A. E., Riedl, K. M., Jones, G. A., Sovik, K. N., Ritchard, N. T., Hartzfeld, P. W., & Riechel, T. L. (1998). High Molecular Weight Plant Polyphenolics (Tannins) as Biological Antioxidants. Journal of Agricultural and Food Chemistry, 46(5), 1887-1892. Laguerre, M., Lopez-Giraldo, L. J., Lecomte, J., Barea, B., Cambon, E., Tchobo, P. F., Barouh, N., & Villeneuve, P. (2008). Conjugated autoxidizable triene (CAT) assay: A novel spectrophotometric method for determination of antioxidant capacity using triacylglycerol as ultraviolet probe. Analytical Biochemistry, 380(2), 282-290. 15
372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422
McClements, D. J. (2007). Critical reviews of techniques and methodologies for characterisation of emulsion stability. Critical review in food science and nutrition, 47(7), 611-647. Michodjehoun-Mestres, L., Souquet, J. M., Fulcrand, H., Meudec, E., Reynes, M., & Brillouet, J. M. (2009). Characterisation of highly polymerised prodelphinidins from skin and flesh of four cashew apple (Anacardium occidentale L.) genotypes. Food Chemistry, 114(3), 989995. Mouls, L., & Fulcrand, H. (2012). UPLC-ESI-MS study of the oxidation markers released from tannin depolymerization: toward a better characterization of the tannin evolution over food and beverage processing. Journal of Mass Spectrometry, 47(11), 1450-1457. Novales, B., Papineau, P., Sire, A., & Axelos, M. A. V. (2003). Characterization of emulsions and suspensions by video image analysis. Colloids and Surfaces A: Physicochem. Eng. Aspects, 221, 81-89. Pazos, M., Gallardo, J. M., Torres, J. L., & Medina, I. (2005). Activity of grape polyphenols as inhibitors of the oxydation of fish lipids and frozen fish muscle. Food Chemistry, 92, 547557. Pickering, S. U. (1907). Emulsions. Journal of the Chemical Society, Transactions, 91(0), 20012021. Poncet-Legrand, C., Cabane, B., Bautista-Ortín, A.-B., Carrillo, S., Fulcrand, H., Pérez, J., & Vernhet, A. (2010). Tannin Oxidation: Intra vs. Intermolecular reactions. Biomacromolecules, 11(9), 2376-2386. Poncet-Legrand, C., Cartalade, D., Putaux, J. L., Cheynier, V., & Vernhet, A. (2003). Flavan-3-ol aggregation in model ethanolic solutions: incidence of polyphenol structure, concentration, ethanol content, and ionic strength. Langmuir, 19, 10563-10572. Porter, W. L., Black, E. D., & Drolet, A. M. (1989). Use of Polyamide Oxidative Fluorescence Test on Lipid Emulsions - Contrast in Relative Effectiveness of Antioxidants in Bulk Versus Dispersed Systems. Journal of Agricultural and Food Chemistry, 37(3), 615-624. Preys, S., Mazerolles, G., Courcoux, P., Samson, A., Fischer, U., Hanafi, A., Bertrand, D., & Cheynier, V. (2006). Relationship between polyphenolic composition and some sensory properties in red wines using multiway analyses. Analytica Chimica Acta, 563(1-2), 126136. Prieur, C., Rigaud, J., Cheynier, V., & Moutounet, M. (1994). Oligomeric and polymeric procyanidins from grape seeds. Phytochemistry, 36(3), 781-784. Slabbert, N. P. (1977). Ionization of Some Flavanols and Dihydroflavonols. Tetrahedron, 33(7), 821-824. Tcholakova, S., Denkov, N.D., Sidzhakova, D., Campbell, B. (2006). Effect of thermal treatment, ionic strength, and pH on the short-term and long-term coamescence stability of βlactoglobulin emulsions. Langmuir, 22, 6042-6052. Tcholakova, S., Denkov,N.D., Sidzhakova,D., Ivanov,I.V., Campbell, B. (2005). Effect of electrolyte concentration and pH on the coalescence stability of β-lactoglobulin emulsions:experiment and interpretation. Langmuir, 21, 4842-4855. Torres, J. L., Varela, B., Garcia, M.T., Carilla, J., Matito, C., , Centelles, J. J., Cascante, M., Xavier, S., & Bobet, R. (2002). Valorization of grape (Vitis vinifera) by-products. Antioxydant and biological properties of polyphenolic fractions differing in procyandin composition and flavanol content. Journal of Agricultural and Food Chemistry, 50, 7548-7555. Van Acker, S. A. B. E., Van Den Berg, D.-j., Tromp, M. N. J. L., Griffioen, D. H., Van Bennekom, W. P., Van Der Vijgh, W. J. F., & Bast, A. (1996). Structural aspects of antioxidant activity of flavonoids. Free Radical Biology and Medicine, 20(3), 331-342. Vernhet, A., Dubascoux, S., Cabane, B., Fulcrand, H., Dubreucq, E., & Poncet-Legrand, C. (2011). Characterization of oxidized tannins: comparison of depolymerization methods, asymmetric flow field-flow fractionation and small-angle X-ray scattering. Analytical and Bioanalytical Chemistry, 401(5), 1559-1569.
16
423 424 425 426 427 428 429 430 431 432 433
Vidal, S., Francis, L., Guyot, S., Marnet, N., Kwiatkowski, M., Gawel, R., Cheynier, V., & Waters, E. J. (2003). The mouth-feel properties of grape and apple proanthocyanidins in a wine-like medium. Journal of the Science of Food and Agriculture, 83(6), 564-573. Wright, J. S., Johnson, E. R., & DiLabio, G. A. (2001). Predicting the Activity of Phenolic Antioxidants: Theoretical Method, Analysis of Substituent Effects, and Application to Major Families of Antioxidants. Journal of the American Chemical Society, 123(6), 11731183. Zanchi, D., Vernhet, A., Poncet-Legrand, C., Cartalade, D., Tribet, C., Schweins, R., & Cabane, B. (2007). Colloidal dispersions of tannins in water-ethanol solutions. Langmuir, 23(20), 99499959.
17
Tables Table 1: Doehlert experimental design for three variables Experiment number
Tannin concentration
pH
Ionic strength (mM)
(% w/w aqueous phase) 1
0.3
6
75
2
0.5
6
75
3
0.4
9
75
4
0.4
7
100
5
0.1
6
75
6
0.2
5
50
7
0.4
5
50
8
0.3
8
50
9
0.2
9
75
10
0.2
7
100
11
0.3
4
100
12
0.3
6
75
13
0.3
6
75
18
Table 2: Experimental design results for grape seeds tannins G15ox. Parameter
aia
etb
tc
p-vald
a0 = -7.592
2.932
-2.589
0.049
TCe
X1
a1= -4.250
5.989
-0.710
0.510
pH
X2
a2 = 1.196
0.428
2.791
0.038
If
X3
a3 = 0.087
0.048
1.810
0.130
TC pH
X1 X2
a12 = 0.250
0.474
0.528
0.620
TC I
X1 X3
a13 = 0.060
0.060
1.001
0.363
pH I
X2 X3
a23 = -0.040
0.003
-11.559
S0308-8146(15)00058-8 http://dx.doi.org/10.1016/j.foodchem.2015.01.056 FOCH 17007
To appear in:
Food Chemistry
Received Date: Revised Date: Accepted Date:
30 September 2014 7 January 2015 8 January 2015
Please cite this article as: Figueroa-Espinoza, M.C., Zafimahova, A., Maldonado Alvarado, P.G., Dubreucq, E., Poncet-Legrand, C., Grape seed and apple tannins: emulsifying and antioxidant properties, Food Chemistry (2015), doi: http://dx.doi.org/10.1016/j.foodchem.2015.01.056
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
Grape seed and apple tannins: emulsifying and antioxidant properties †
2
Maria Cruz Figueroa-Espinoza1, Andrea Zafimahova1,2,3,4, Pedro G. Maldonado Alvarado1,2,3,4,
3
Eric Dubreucq1, Céline Poncet-Legrand2,3,4,*
4 5 6 1
7 8
2
Montpellier SupAgro, UMR1208 IATE, F-34060 Montpellier, France
INRA, 3 Montpellier SupAgro, 4 Université Montpellier II, UMR1083 SPO, F-34060 Montpellier,
9
France
10 11 12 13
*
Corresponding author: [email protected]
14 15 16 17
† Dedicated to the memory of Gérard Mazerolles
18
1
19
Abstract
20 21
Tannins are natural antioxidants found in plant-based foods and beverages, whose amphiphilic
22
nature could be useful to both stabilize emulsions and protect unsaturated lipids from oxidation. In
23
this paper, the use of tannins as antioxidant emulsifiers was studied. The main parameters
24
influencing the stability of emulsions (i.e. tannins structure and concentration, aqueous phase pH,
25
and ionic strength) were identified and optimized. Oil in water emulsions stabilized with tannins
26
were compared with those stabilized with two commercial emulsifying agents, poly(vinyl alcohol)
27
(PVA) and polyoxyethylene hydrogenated castor oil. In optimized conditions, the condensed
28
tannins allowed to obtain a stability equivalent to that of PVA. Tannins presented good antioxidant
29
activity in oil in water emulsion, as measured by the Conjugated Autoxidizable Triene (CAT)
30
assay.
31 32
Keywords: antioxidant, polyphenols, tannins, catechin, emulsion, CAT assay, poly(vinyl alcohol).
33 34
2
35
1. Introduction
36
Emulsions are formed when two immiscible liquids are mixed. One liquid (the dispersed
37
phase) is dispersed as droplets (from 0.10 µm to a few µ m of diameter) in the other (the continuous
38
phase) (McClements, 2007). Forming emulsions requires energy to increase the interfacial area
39
between continuous and dispersed phases. Emulsified systems are thermodynamically unstable
40
because of the surface tension between oil and water, which opposes to the increase of interfacial
41
area. Emulsions can be stabilized by amphiphilic molecules, which adsorb in the oil-water
42
interface, decreasing the surface tension. They can also be stabilized by solid particles which adsorb
43
onto the interface between the two phases and are called Pickering emulsions (Pickering, 1907).
44
Besides, the stability of emulsions is conditioned by the competition between attractive (Van der
45
Waals, hydrophobic interactions, electrostatic attractions, hydrogen bonds) and repulsive forces
46
(electrostatic repulsion, steric repulsion) between the dispersed droplets (Guzey & McClements,
47
2007; Tcholakova, Denkov, Sidzhakova, Ivanov, & Campbell, 2005). Stabilizing agents increase
48
repulsions between droplets and slow down coalescence and phase separation phenomena. The
49
stabilization of emulsions also depends on constituents of the emulsion, namely the concentration in
50
emulsifier or stabilizing agent, pH, viscosity and ionic strength of the aqueous phase, and the
51
concentration of the organic phase (Chanamai & McClements, 2000; Tcholakova, Denkov,
52
Sidzhakova, & Campbell, 2006). Emulsified systems raise the problem of oxidation: dispersal of
53
lipids in emulsified systems increases the specific area in contact with oxygen and some pro-
54
oxidizing species (Coupland & McClements, 1996), which are detrimental to lipids.
55
On another hand, polyphenols are known for their antioxidant properties. They are molecules found
56
in large amounts in plant-based foods and beverages. They are constituted by one or several rather
57
hydrophobic aromatic nuclei bearing polar hydroxyl groups (-OH) (Poncet-Legrand, Cartalade,
58
Putaux, Cheynier, & Vernhet, 2003). Among them, a particular class of polyphenols called tannins
59
is produced in large amounts by distilleries and wine industries. These compounds are polymers of
3
60
flavan-3-ol units primarily linked by C4-C8 bonds, with C4-C6 bonds giving rise to some degree of
61
polymer branching (Figure 1). For example, the constitutive units of grape seed tannins are (+)-
62
catechin (C), (-)-epicatechin (Ec), and (-)-epicatechin gallate (EcG) (Prieur, Rigaud, Cheynier, &
63
Moutounet, 1994). Conversely, apple condensed tannins present an homogeneous structure, with
64
primarily (-)-epicatechin as the constitutive flavanol unit (Guyot, Doco, Souquet, Moutounet, &
65
Drilleau, 1997). The antibacterial tannin activity is related to surface chemistry and to their ability
66
to non-covalently associate with proteins and other macromolecular structures (Vidal, Francis,
67
Guyot, Marnet, Kwiatkowski, Gawel, et al., 2003).
68
Their structure (aromatic hydrophobic rings, hydroxyl hydrophilic groups) suggests that some
69
tannins may have surface active properties, and might thus stabilize emulsions. Although there are
70
already numerous natural surfactant agents, the interest to use tannins as stabilizers lies in the fact
71
that they also have important antioxidant capacities. They are highly polymerized and possess many
72
phenolic hydroxyl groups. In addition, the “B” ring in the flavanols is responsible for most of the
73
antioxidant activity, as it contains the catechol or trihydroxy functionality. (Hagerman, Riedl,
74
Jones, Sovik, Ritchard, Hartzfeld, et al., 1998; Pazos, Gallardo, Torres, & Medina, 2005; Torres,
75
Centelles, Cascante, Xavier, & Bobet, 2002; Van Acker, Van Den Berg, Tromp, Griffioen, Van
76
Bennekom, Van Der Vijgh, et al., 1996). However, due to their high chemical reactivity, they
77
undergo numerous reactions when they are in solution, leading to the formation of new compounds
78
which possess chemical structures, conformation in solution, and water solubility different from
79
those of native tannins.
80
In this paper, the following properties of catechin, grape seed and apple tannins were studied: oil in
81
water emulsion stabilization capacity and antioxidant activity in emulsified system. The effect of
82
the tannin’s structure (chemical composition and polymerization degree) and concentration, as well
83
as of pH and ionic strength on the emulsion stability was also investigated.
84
4
85
2. Materials and Methods
86
2.1.
Materials
87
Methyl oleate (the oil phase) and Eumulgin ® HRE40, an ether of polyoxyethylene
88
cetyl stearyl alcohol, were supplied by Cognis (Monheim, Germany; now BASF). Phosphate
89
buffers (pH 3 to 9, ionic strengths 50 to 100 mM) were prepared in deionised water (MilliQ system,
90
Millipore, USA) from reagents obtained from Sigma Aldrich (St. Louis, MO, USA). Poly(vinyl
91
alcohol) (PVA) was purchased from Merck (Whitehouse Station, NJ, USA).
92
In this paper, tannins will be referred to as Gn or An (repectively for Grape and Apple),
93
where n stands for the number average degree of polymerization, followed by the suffix “ox” when
94
they were oxidized. The polyphenols used in this study were a commercial monomer (catechin,
95
DP1) supplied by Sigma (St. Louis, MO), tannin fractions purified from Grape seed (Vitis vinifera,
96
var. Shiraz) (G4ox and G15ox) obtained as referred in the literature (Poncet-Legrand, Cartalade,
97
Putaux, Cheynier, & Vernhet, 2003), and Apple (Malus sylvestris var. Kermerrien) (A6 and 15) as
98
in reference (Michodjehoun-Mestres, Souquet, Fulcrand, Meudec, Reynes, & Brillouet, 2009).
99
Their degree of polymerization was determined by depolymerisation followed by HPLC analysis,
100
as described by Preys et al. (Preys, Mazerolles, Courcoux, Samson, Fischer, Hanafi, et al., 2006)
101
immediately after their purification. Depolymerisation was also performed prior to emulsion
102
preparation, and we observed a drop in depolymerisation yield compared to the initial one with the
103
grape seed fractions (but not the apple tannins) and this was attributed to oxidation. In this case, the
104
characterization of fractions by standards methods becomes inaccurate (Mouls & Fulcrand, 2012;
105
Poncet-Legrand, Cabane, Bautista-Ortín, Carrillo, Fulcrand, Pérez, et al., 2010; Vernhet,
106
Dubascoux, Cabane, Fulcrand, Dubreucq, & Poncet-Legrand, 2011): the effective degrees of
107
polymerization of these oxidized tannins is larger. In a second set of experiments, emulsions were
108
prepared with freshly purified apple tannins (A15b). Oxidized catechin (D1ox) was obtained by
109
stocking a catechin (D1) solution (3 mg mL-1 in phosphate buffer pH 6.5, 50 mM) at 28°C for 7
110
days. To obtain oxidized apple tannins (A15box), 33 mg of apple tannins (A15b) were solubilized 5
111
in 11 mL of filtered (0.22 µm) phosphate buffer (pH 6.5, 50 mM), and left in contact with air
112
during 15 days at 28 °C. The buffers were filtered in order to minimize the risks of bacterial
113
growth. Oxidized tannins exhibit a different color from the non-oxidized ones: their color in
114
solution is ranging from dark yellow to brown. Depending on the application (food or non-food
115
application) these color changes may or may not be accepted from a consumer point of view.
116
2.2.
117
Methods 2.2.1. Emulsion preparation
118
Tannins were dissolved in the adequate phosphate buffer at concentrations ranging from
119
0.1 to 1 % (w/w), 24 hours prior to emulsion preparation. Methyl oleate (10 % v/v of the total
120
emulsion volume) was then added to the aqueous phase. Preparation of the emulsion was performed
121
by sonicating the mixture during 30 s at room temperature (20°C) using an output signal amplitude
122
of 15 µm (peak to peak), placing the probe sonicator (Sonifier 250, Branson Ultrasonics Co., Paris,
123
France) vertically in the centre of the recipient, about 5 mm from the bottom.
124
2.2.2. Emulsion properties
125
Conductivity measurements were done using a conductimeter to determine whether the
126
emulsions were direct (the continuous phase is water; high conductivity) or inverse (the continuous
127
phase is oil; zero conductivity). The granulometric analysis of emulsions was done with a light
128
scattering instrument (Mastersizer 2000 laser granulometer, Malvern, Malvern, UK). Stability was
129
monitored over time by measuring the emulsified volume (Ve) compared to the total volume (Vt).
130
2.2.3. Antioxidant properties of tannins
131
Antioxidant capacity of tannin fractions was measured using the CAT (conjugated
132
autoxidizable triene) procedure developed by Laguerre et al. (Laguerre, Lopez-Giraldo, Lecomte,
133
Barea, Cambon, Tchobo, et al., 2008). Briefly, this assay is based upon the high sensitivity to
134
oxidation of the α-eleostearic acid present in the triacylglycerols of Tung (Vernicia fordii) oil.
6
135
Eleostearic acid (ELA) is a linolenic acid containing a conjugated triene part: it thus has a strong
136
UV absorbance at λmax = 273 nm. Upon oxidation, the degradation of the conjugated triene system
137
into a conjugated diene is accompanied by a decrease in the signal at 273 nm (and an increase at
138
234 nm). Tung oil oxidation was initiated by 2,2’-azobis(2-amidinopropane) dihydrochloride
139
(AAPH) in a water in oil emulsion at 37 °C, in the presence or absence of the tested antioxidant.
140
Spectrophotometric monitoring of the UV signal decay at 273 nm determines the ability of a
141
molecule to protect tung oil from oxidation. Trolox (6-hydroxy-2,5,7,8-tetramethylchroman-2-
142
carboxylic acid) was used as reference antioxidant. The CAT value is expressed as micromoles of
143
trolox equivalents (TE) per gram of dried tannins.
144
2.2.4. Experimental design to optimize emulsion stability
145
Emulsion stability was monitored overtime by measuring the emulsified volume (Ve) with regards
146
to the total volume (Vt) (McClements, 2007; Novales, Papineau, Sire, & Axelos, 2003). A Doehlert
147
experimental design (Doehlert, 1970) was built for determining the experimental conditions that
148
provide the highest emulsion volume using three factors (pH 4.0-9.0; tannin concentration (0.1-0.5
149
% w/w), ionic strength (50-100 mM)) at 7, 6, and 5 levels respectively. A total of 13 experiments
150
were realized, including three replicates at the central point (Table 1). The total amount of
151
emulsion was 10 mL, prepared with 10 % of oil.
152
The model to which the experimental data were fitted was a second-order polynomial
153
model. The following equation was used:
154
Y = a0 + a1 X1 + a2 X2 + a3 X3 + a12 X1X2 + a13 X1X3 + a23 X2X3 + a11 X12 + a22 X22 + a33 X32
155
where Y is the emulsified volume, ai are the regression coefficients, and X1, X2, and X3 are the
156
experimental factors. Confidence intervals were evaluated with T95%,3 as the Student coefficient.
157
7
158
3. Results and discussion
159
Tannins present a huge diversity of structures (monomer composition, degree of polymerization
160
(DP), branching). Prior to the optimization of pH and ionic strength, two important parameters in
161
emulsion stabilization (Chanamai & McClements, 2000; Gu, Decker, & McClements 2004), a
162
preliminary study of the effect of the polymer structure on emulsion stability was performed. The
163
emulsifying properties of catechin (DP1) and oxidized catechin (DP1ox), two fractions of grape
164
seeds tannins (DP4 and 15, named G4ox and G15ox) and of two fractions of apple tannins (DP6
165
and 15, named A6 and A15), were compared. Indeed apple and grape seed tannins differ in
166
composition: apple tannins are homopolymers of epicatechin, while grape seed tannins are random
167
copolymers of catechin, epicatechin and epicatechin gallate (Figure 1).
168
After determining which tannins were the best stabilizers, tannin concentration, pH, and
169
ionic strength, were optimized and a comparison was made with two commercial emulsifiers (PVA
170
and Eumulgin HRE40).
171 172
3.1.
173
degree)
Effect of tannin structure (chemical composition, degree of polymerization, oxidation
174
The effect of tannin structure was studied by preparing emulsions with catechin
175
(monomer), apple and grape seed tannin fractions having a different DP’s. Oil in water emulsions
176
were prepared in phosphate buffer (50 mM, pH 7.0) with 10 % of methyl oleate, and emulsifier
177
(0.1 % w/v of the water phase). After emulsification by sonication, the turbidity of the emulsion in
178
the transparent tube was the same on all the height of the tube. Emulsion destabilization was then
179
monitored during at least 3 days at room temperature and the ratio emulsified volume/total volume
180
was plotted as a function of time (Figure 2).
181
A phase separation occurred very quickly (within two minutes) for reference emulsions (i.e. methyl
182
oleate and buffer without added emulsifier). The same behavior was observed for the emulsions 8
183
stabilized with catechin (DP1) and oxidized catechin (DP1ox), not shown, as well as apple tannins,
184
except for A6, which provided a good stabilisation during the first hour. After 24 hours, emulsified
185
volumes obtained with Eumulgin were superior to those obtained with the PVA (97 % vs 75 %) and
186
grape seed tannins. For these two last ones, emulsified volumes were higher with G15ox (95 %)
187
than with G4ox (30 %) after 24 hours. These values were the same after 3 days, except for G15ox:
188
emulsion represented 80 % of the volume. Grape seeds tannins with the highest DPs (G15ox)
189
provided the best stabilization.
190
Tannins are thus able to stabilize emulsions under certain conditions. In the conditions
191
tested, their chemical structures had a strong impact on their physicochemical properties: grape
192
tannins stabilizing properties increased with their DP, and for a given DP, galloylated tannins (i.e.
193
grape seeds tannins, having ~ 25 % (w/w) of epicatechin gallate units) seemed to be more efficient
194
than apple tannins.
195
However, the grape seed tannin fraction used was oxidized and it was thus difficult to
196
compare apple and grape seed tannins. Upon oxidation, tannins undergo many chemical reactions
197
and some of these reactions lead to the formation of higher molecular weight species (Poncet-
198
Legrand, Cabane, Bautista-Ortín, Carrillo, Fulcrand, Pérez, & Vernhet, 2010; Vernhet, Dubascoux,
199
Cabane, Fulcrand, Dubreucq, & Poncet-Legrand, 2011). Furthermore, the conformation of the
200
macromolecules changes (e.g. from linear to branched macromolecules or from semi-flexible to
201
more rigid chains), and water-solubility of the fractions also changes. Analysis of oxidized tannins
202
(real DP, conformation in solution) is not easy and when analyzing G15ox by Small Angle X-Ray
203
Scattering (Vernhet, Dubascoux, Cabane, Fulcrand, Dubreucq, & Poncet-Legrand, 2011), results
204
showed that the actual degree of polymerization of grape seed tannin was 25 rather than 15. The
205
impact of oxidation was confirmed by preparing emulsions with apple tannins (A15b) just after
206
they were extracted and purified, and after oxidation in phosphate buffer 50 mM, pH 6.5 (A15box):
207
unoxidized tannins did not stabilize emulsions, whereas the oxidized ones did (results not shown).
9
208
Emulsion stabilizers are often surface active agents that are anchored at the interface
209
between oil and aqueous solution to reduce surface tension. It was thus verified if tannins were
210
surface active by measuring surface tension of tannin solution: at the concentration used in this
211
study, the surface tension was not significantly decreased, meaning that in our conditions (ionic
212
strength, pH, concentration), the tannins used in this study were not surface active (Supplementary
213
data S1).
214
The mechanism of emulsion stabilization by tannins may be the result of steric and/or
215
electrostatic repulsions between droplets covered by adsorbed polymers. Last, but not least,
216
emulsions can also be stabilized by particles (Pickering, 1907), and tannins, especially the oxidized
217
ones, have been reported to form aggregates and particles (Poncet-Legrand, Cartalade, Putaux,
218
Cheynier, & Vernhet, 2003; Zanchi, Vernhet, Poncet-Legrand, Cartalade, Tribet, Schweins, et al.,
219
2007). The stabilization increased with the tannin degree of polymerization (which remained low in
220
comparison with other types of biopolymers: all tannins used in this study had molecular weight
0.5). pH was the parameter which had more effect on the
242
emulsified volume, as well as its combination with itself and the ionic strength. It is worth noting
243
that the quadratic form of I has also an effect (p values < 0.05).
244
Prediction profiles of the emulsion stability for pH 4, 6, and 8 are shown on Figure 3. At
245
pH 4 (Figure 3A), the emulsified volume increased very slightly with tannin concentration and
246
ionic strength. It is thus likely that the emulsion stability is governed by the steric repulsion
247
between droplets, due to the multilayer tannin adsorption. However, at pH values ranging from 6 to
248
8 (Figure 3A and B), tannin concentration effect is limited and the emulsified volume decreases
249
when the ionic strength increases. Maximum of emulsified volumes would be obtained for values
250
of pH superior to 8 and ionic strengths lower than 50 mM. In this pH range, tannins are more and
251
more negatively charged: catechin and epicatechin pKas are 8.2 and 9.2, respectively (Slabbert,
252
1977). Droplets are mainly stabilized by electrostatic repulsions. Increasing ionic strength at these
253
pHs induce the screening of the charges and thus a lowering of the repulsions. At lower pH (4 for
254
instance), tannins are not charged, and when ionic strength is increasing, they form particles more
255
easily due to hydrophobic interactions (Poncet-Legrand, Cartalade, Putaux, Cheynier, & Vernhet,
256
2003). In the hypothesis of emulsions stabilized by tannin aggregates (Pickering emulsion), this
257
would explain the improved stabilization observed when the ionic strength is increased.
11
258 259
3.3.
Emulsion ageing
260
The effect of grape seeds tannins G15ox on the emulsion ageing was performed by
261
granulometric analysis by monitoring the emulsified volume decrease as a function of time
262
(Supplementary data S2 and S3). Emulsions contained either 50 % (w/w) or 10 % (w/w) of methyl
263
oleate in phosphate buffer (50 mM, pH 7) and tannins G15 (0.9 % w/v of the oil phase). A “fresh”
264
emulsion (one hour old) and a one month old emulsion were compared. In the first experiment
265
(50% of methyl oleate), 10 % of the formed droplets had a size of the order of 0.7 µm. About 50 %
266
of droplets had a size lower or equal to 1 µm and 10 % had a diameter superior to 2.2 µm. After a
267
month of ageing, 10 % of the formed droplets had a size of the order of 2 µm. About 50 % of
268
droplets had a size lower or equal to 3 µm and 10 % had a diameter superior to 5.6 µm. The
269
droplets size thus roughly tripled during the storage.
270
In the second experiment (10% of methyl oleate), less than 10 % of the formed droplets
271
had a size of the order of 0.8 µm. About 50 % of droplets had a size lower or equal to 1.5 µm and
272
10 % had a diameter superior to 3.5 µm. After a month of ageing, the drops average size was about
273
4.6 µm, with a maximal size around 9.2 µm for 10 % of droplets and a minimal size of 3 µm. The
274
droplets size also tripled during the storage.
275
Such a diameter increase may be due to flocculation or to coalescence. In the first case,
276
it would mean that the emulsion degradation is reversible, and that after a certain excitement the
277
emulsion would take back the initial droplets size distribution. In the second case, the degradation
278
is irreversible and it would be necessary to re-emulsify the mixture. Microscopy observations
279
(supplementary data S4) showed that the size of the primary droplets did not increase, whereas
280
aggregates are getting larger after one month. Flocculation is thus the driving mechanism for
281
emulsion destabilization in our systems.
12
282
3.4.
Antioxidant properties
283
Antioxidant properties of different tannins was evaluated by the use of the CAT assay
284
(Laguerre, et al., 2008) using triacylglycerols of tung oil as an ultraviolet probe in an emulsified
285
medium and results are presented in Figure 4. As an example, the kinetics of bleaching in the
286
absence of oxidized apple tannins A15ox in weight equivalents is illustrated. Trolox and all tested
287
grape seed and apple tannins delayed AAPH induced oxidation of stripped tung oil, without
288
exhibiting a significant lag phase. As explained by Laguerre et al. (Laguerre, et al., 2008) this
289
absence of a lag phase is characteristic of molecules that rather scavenge AAPH-derived peroxyl
290
radicals, instead of directly reducing the lipoperoxyl radical derived from oil. Thus Trolox and
291
tannins seem to behave as retarder antioxidants rather than pure chain-breaking antioxidants.
292
Tannins are fairly good antioxidant: grape seed tannins G15ox presented a CAT value of 7200 ±
293
390 TE/ g, followed by non oxidized apple tannins A6 (5490 ± 320 TE/g) and oxidized apple
294
tannins A15ox (3850 ± 430 TE/g). Our results are in the same order of magnitude with the results
295
of previous reported results of 8200 TE/g of (-)-epicatechin and 2530 TE/g of gallic acid
296
(calculated from reference (Laguerre, et al., 2008)). Considering that the location of phenolics
297
toward the oil droplet in an oil in water emulsion is of prime importance in their antioxidant
298
capacities, such a high CAT value for G15ox tannins reflects their good location toward the
299
oxidation site, in this case the lipidic substrate, according to the polar paradox (Porter, Black, &
300
Drolet, 1989). These results are supported by those obtained in section 3.1 showing that grape seed
301
tannins G15ox were the best emulsion stabilizers. Thus, G15ox capacity of adsorbing into the
302
interface may contribute to increase their concentration at the site where oxidation occurs. Besides
303
this, grape seeds tannins, unlike apple tannins whose aromatic rings are dihydroxylated, contain
304
galloylated units with a trihydroxylation pattern. This different hydroxylation substitution, in
305
addition to their high molecular weight and the proximity of many aromatic rings and hydroxyl
306
groups, contribute to increase their antioxidant capacity. The order of antioxidant activity has been
307
reported as (-)-epicatechin < epicatechin-3-O-gallate < (-)-epigallocatechin (Wright, Johnson, &
13
308
DiLabio, 2001), which support our results, grape seed tannins G15ox containing both, (-)-
309
epicatechin and epicatechin gallate units, and apple tannins (A6 and A15) containing only (-)-
310
epicatechin units.
311
Tannins antioxidant properties come from their hydroxyl groups and more specifically
312
from their catechol moieties (i.e. –OH groups on ortho positions on aromatic rings). Grape seed
313
tannins are galloylated and contain more catechol units than the apple ones, which may explain that
314
they are also more antioxidant. When oxidation takes place, it does not lead to the destruction of
315
catechol moieties (S. Guyot, Vercauteren, & Cheynier, 1996) (at least not systematically),
316
explaining why oxidized apple tannins A15ox still have antioxidant activity. This has practical
317
implications, since tannin fractions obtained from distilleries are rarely unoxidized. It is important
318
that they still have antioxidant properties.
319
320
4. Conclusion
321
In this project we investigated the use of phenolic compounds obtained from grape seed
322
and apple as emulsion and antioxidant stabilizers. This first study showed that oxidized grape seed
323
tannins can be used to stabilize oil in water emulsion. Tannin structure plays an important part on
324
emulsion properties: best candidates are relatively high molecular weight tannins, which underwent
325
oxidation reactions (i.e. their chemical structure, and thus their solution properties are different
326
from initial molecules). Emulsions stabilized with tannins were compared with emulsions obtained
327
in the same conditions using model emulsifiers as PVA and with Eumulgin®. In certain
328
physicochemical conditions, the oxidized tannins allowed to obtain a stability equivalent to that of
329
the PVA. The oil in water formed emulsions were up to 50 % of methyl oleate, with drop mean
330
sizes in the micrometer range. Emulsion stabilization appeared to be due to electrostatic repulsions
331
at high pH’s (~ pKa), and emulsion breaking was mainly due to droplet flocculation, as observed by
332
laser granulometry experiments. At pH A6 > A15ox)
337
measured by the CAT test. This research shows the potential use of the winery and distillery by-
338
products and wastes for the production of high added value extracts rich in oxidized tannins that
339
could be used as antioxidants emulsifiers.
340
341 342
Acknowledgements This work was financed by Montpellier SupAgro. We thank Gérard Mazerolles who
343
helped us in the design of experiment and in the interpretation of its results.
344
References
345 346 347 348 349 350 351 352 353 354 355 356 357 358 359 360 361 362 363 364 365 366 367 368 369 370 371
Cartalade, D., & Vernhet, A. (2006). Polar interactions in flavan-3-ol adsorption on solid surfaces. Journal of Agricultural and Food Chemistry, 54(8), 3086-3094. Chanamai, R. & McClements, D.J. (2000). Dependance of creaming and rheology of monodisperse oil-in-water emulsions on droplet size and concentration. Colloids and Surfaces, 172, 79-86. Coupland, J. N., & McClements, D. J. (1996). Lipid oxydation in food emulsions. Trends in Food Science and technology, 7, 83-91. Doehlert, D. H. (1970). Uniform Shell Designs. Journal of the Royal Statistical Society. Series C (Applied Statistics, 19(3), 231-239. Gu, Y. S., Decker, E.A., & McClements, D.J. (2004). Influence of pH and carrageenan type on properties of β-lactoglobulin stabilized oil-in water emulsions. Food Hydrocolloids, 19, 8391. Guyot, S., Doco, T., Souquet, J.-M., Moutounet, M., & Drilleau, J.-F. (1997). Characterization of highly polymerized procyanidins in cider apple (Malus sylvestris var. kermerrien) skin and pulp. Phytochemistry, 44(2), 351-357. Guyot, S., Vercauteren, J., & Cheynier, V. (1996). Colourless and yellow dimers resulting from (+)catechin oxidative coupling catalysed by grape polyphenoloxidase. Phytochemistry, 42, 1279-1288. Guzey, D., & McClements, D. J. (2007). Impact of electrostatic interactions on formation and stability of emulsions containing oil droplets coated by β-lactoglobulin-Pectin complex. Journal of Agricultural and Food Chemistry, 55, 475-485. Hagerman, A. E., Riedl, K. M., Jones, G. A., Sovik, K. N., Ritchard, N. T., Hartzfeld, P. W., & Riechel, T. L. (1998). High Molecular Weight Plant Polyphenolics (Tannins) as Biological Antioxidants. Journal of Agricultural and Food Chemistry, 46(5), 1887-1892. Laguerre, M., Lopez-Giraldo, L. J., Lecomte, J., Barea, B., Cambon, E., Tchobo, P. F., Barouh, N., & Villeneuve, P. (2008). Conjugated autoxidizable triene (CAT) assay: A novel spectrophotometric method for determination of antioxidant capacity using triacylglycerol as ultraviolet probe. Analytical Biochemistry, 380(2), 282-290. 15
372 373 374 375 376 377 378 379 380 381 382 383 384 385 386 387 388 389 390 391 392 393 394 395 396 397 398 399 400 401 402 403 404 405 406 407 408 409 410 411 412 413 414 415 416 417 418 419 420 421 422
McClements, D. J. (2007). Critical reviews of techniques and methodologies for characterisation of emulsion stability. Critical review in food science and nutrition, 47(7), 611-647. Michodjehoun-Mestres, L., Souquet, J. M., Fulcrand, H., Meudec, E., Reynes, M., & Brillouet, J. M. (2009). Characterisation of highly polymerised prodelphinidins from skin and flesh of four cashew apple (Anacardium occidentale L.) genotypes. Food Chemistry, 114(3), 989995. Mouls, L., & Fulcrand, H. (2012). UPLC-ESI-MS study of the oxidation markers released from tannin depolymerization: toward a better characterization of the tannin evolution over food and beverage processing. Journal of Mass Spectrometry, 47(11), 1450-1457. Novales, B., Papineau, P., Sire, A., & Axelos, M. A. V. (2003). Characterization of emulsions and suspensions by video image analysis. Colloids and Surfaces A: Physicochem. Eng. Aspects, 221, 81-89. Pazos, M., Gallardo, J. M., Torres, J. L., & Medina, I. (2005). Activity of grape polyphenols as inhibitors of the oxydation of fish lipids and frozen fish muscle. Food Chemistry, 92, 547557. Pickering, S. U. (1907). Emulsions. Journal of the Chemical Society, Transactions, 91(0), 20012021. Poncet-Legrand, C., Cabane, B., Bautista-Ortín, A.-B., Carrillo, S., Fulcrand, H., Pérez, J., & Vernhet, A. (2010). Tannin Oxidation: Intra vs. Intermolecular reactions. Biomacromolecules, 11(9), 2376-2386. Poncet-Legrand, C., Cartalade, D., Putaux, J. L., Cheynier, V., & Vernhet, A. (2003). Flavan-3-ol aggregation in model ethanolic solutions: incidence of polyphenol structure, concentration, ethanol content, and ionic strength. Langmuir, 19, 10563-10572. Porter, W. L., Black, E. D., & Drolet, A. M. (1989). Use of Polyamide Oxidative Fluorescence Test on Lipid Emulsions - Contrast in Relative Effectiveness of Antioxidants in Bulk Versus Dispersed Systems. Journal of Agricultural and Food Chemistry, 37(3), 615-624. Preys, S., Mazerolles, G., Courcoux, P., Samson, A., Fischer, U., Hanafi, A., Bertrand, D., & Cheynier, V. (2006). Relationship between polyphenolic composition and some sensory properties in red wines using multiway analyses. Analytica Chimica Acta, 563(1-2), 126136. Prieur, C., Rigaud, J., Cheynier, V., & Moutounet, M. (1994). Oligomeric and polymeric procyanidins from grape seeds. Phytochemistry, 36(3), 781-784. Slabbert, N. P. (1977). Ionization of Some Flavanols and Dihydroflavonols. Tetrahedron, 33(7), 821-824. Tcholakova, S., Denkov, N.D., Sidzhakova, D., Campbell, B. (2006). Effect of thermal treatment, ionic strength, and pH on the short-term and long-term coamescence stability of βlactoglobulin emulsions. Langmuir, 22, 6042-6052. Tcholakova, S., Denkov,N.D., Sidzhakova,D., Ivanov,I.V., Campbell, B. (2005). Effect of electrolyte concentration and pH on the coalescence stability of β-lactoglobulin emulsions:experiment and interpretation. Langmuir, 21, 4842-4855. Torres, J. L., Varela, B., Garcia, M.T., Carilla, J., Matito, C., , Centelles, J. J., Cascante, M., Xavier, S., & Bobet, R. (2002). Valorization of grape (Vitis vinifera) by-products. Antioxydant and biological properties of polyphenolic fractions differing in procyandin composition and flavanol content. Journal of Agricultural and Food Chemistry, 50, 7548-7555. Van Acker, S. A. B. E., Van Den Berg, D.-j., Tromp, M. N. J. L., Griffioen, D. H., Van Bennekom, W. P., Van Der Vijgh, W. J. F., & Bast, A. (1996). Structural aspects of antioxidant activity of flavonoids. Free Radical Biology and Medicine, 20(3), 331-342. Vernhet, A., Dubascoux, S., Cabane, B., Fulcrand, H., Dubreucq, E., & Poncet-Legrand, C. (2011). Characterization of oxidized tannins: comparison of depolymerization methods, asymmetric flow field-flow fractionation and small-angle X-ray scattering. Analytical and Bioanalytical Chemistry, 401(5), 1559-1569.
16
423 424 425 426 427 428 429 430 431 432 433
Vidal, S., Francis, L., Guyot, S., Marnet, N., Kwiatkowski, M., Gawel, R., Cheynier, V., & Waters, E. J. (2003). The mouth-feel properties of grape and apple proanthocyanidins in a wine-like medium. Journal of the Science of Food and Agriculture, 83(6), 564-573. Wright, J. S., Johnson, E. R., & DiLabio, G. A. (2001). Predicting the Activity of Phenolic Antioxidants: Theoretical Method, Analysis of Substituent Effects, and Application to Major Families of Antioxidants. Journal of the American Chemical Society, 123(6), 11731183. Zanchi, D., Vernhet, A., Poncet-Legrand, C., Cartalade, D., Tribet, C., Schweins, R., & Cabane, B. (2007). Colloidal dispersions of tannins in water-ethanol solutions. Langmuir, 23(20), 99499959.
17
Tables Table 1: Doehlert experimental design for three variables Experiment number
Tannin concentration
pH
Ionic strength (mM)
(% w/w aqueous phase) 1
0.3
6
75
2
0.5
6
75
3
0.4
9
75
4
0.4
7
100
5
0.1
6
75
6
0.2
5
50
7
0.4
5
50
8
0.3
8
50
9
0.2
9
75
10
0.2
7
100
11
0.3
4
100
12
0.3
6
75
13
0.3
6
75
18
Table 2: Experimental design results for grape seeds tannins G15ox. Parameter
aia
etb
tc
p-vald
a0 = -7.592
2.932
-2.589
0.049
TCe
X1
a1= -4.250
5.989
-0.710
0.510
pH
X2
a2 = 1.196
0.428
2.791
0.038
If
X3
a3 = 0.087
0.048
1.810
0.130
TC pH
X1 X2
a12 = 0.250
0.474
0.528
0.620
TC I
X1 X3
a13 = 0.060
0.060
1.001
0.363
pH I
X2 X3
a23 = -0.040
0.003
-11.559
