May 19, 2017 - Where qt is the adsorption capacity at the given contact time t (mg ..... classified into various subclasses, i.e. phlorotannins with phenyl linkages, ether linkages ... (BHT), propyl gallate (PG), Octylgalate (OG) and tertiary butyl ... The authors would like to acknowledge the financial support from AKA-ERNC 009.
Accepted Manuscript Purification of phlorotannins from Macrocystis pyrifera using macroporous resins. A. Leyton, J.R. Vergara-Salinas, J.R. Pérez-Correa, M.E. Lienqueo PII: DOI: Reference:
S0308-8146(17)30918-4 http://dx.doi.org/10.1016/j.foodchem.2017.05.114 FOCH 21175
To appear in:
Food Chemistry
Received Date: Revised Date: Accepted Date:
9 July 2016 19 May 2017 22 May 2017
Please cite this article as: Leyton, A., Vergara-Salinas, J.R., Pérez-Correa, J.R., Lienqueo, M.E., Purification of phlorotannins from Macrocystis pyrifera using macroporous resins., Food Chemistry (2017), doi: http://dx.doi.org/ 10.1016/j.foodchem.2017.05.114
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 2 3 4 5 6 7 8 9 10 11 12 13 14 15 16 17
Purification of phlorotannins from Macrocystis pyrifera using macroporous resins. Leyton A.a,b; Vergara-Salinas J.Rc; Pérez-Correa J.Rd; Lienqueo M.E.a. a
Center for Biotechnology and Bioengineering (CeBiB) University of Chile. Department of Chemical Engineering and Biotechnology, Beauchef 851, Santiago Chile. b
Center for Biotechnology and Bioengineering (CeBiB) Universidad de La Frontera, Temuco.
b
Pontificia Universidad Católica de Valparaíso, Escuela de Ingeniería Bioquímica, Avenida Brasil 2085, Valparaíso, Chile. c
Pontificia Universidad Católica de Chile, Department of Chemical and Bioprocess Engineering, ASIS-UC Interdisciplinary Research Program on Tasty, Safe and Healthy Foods, Avenida Vicuña Mackenna 4860, Santiago, Chile.
18 19 20
Phlorotannins are secondary metabolites produced by brown seaweed, which are
21
known for their nutraceutical and pharmacological properties. The aim of this work was to
22
determine the type of macroporous resin and the conditions of operation that improve the
23
purification of phlorotannins extracted from brown seaweed, Macrocystis pyrifera. For the
24
purification of phlorotannins, six resins (HP-20, SP-850, XAD-7, XAD-16N, XAD-4 and
25
XAD-2) were assessed. The kinetic adsorption allowed determination of an average
26
adsorption time for the resins of 9 hours. The highest level of purification of phlorotannins
27
was obtained with XAD-16N, 42%, with an adsorption capacity of 183 ± 18 mg PGE/g
28
resin, and a desorption ratio of 38.2 ± 7.7%. According to the adsorption isotherm the best
29
temperature of operation was 25°C, and the model that best described the adsorption
30
properties was the Freundlich model. The purification of phlorotannins might expand their
31
use as a bioactive substance in the food, nutraceutical and pharmaceutical industries.
32 33 34
Keywords: Separation, bioactive compounds, brown seaweed.
Abstract
35
1. Introduction
36
In the last decades, the chemistry of natural products of marine origin has been the
37
object of intense research to find new compounds with pharmacological and nutraceutical
38
properties, where seaweed have been identified as the main source of compounds with
39
bioactive properties [Barba, 2016; Roohinejad et al., 2016]. Among the compounds with
40
high added value present in seaweed are phlorotannins, polyphenolic compounds formed by
41
oligomers and polymers of the phloroglucinol monomer [Shibata et al., 2004], which have
42
been recognized as antioxidants, antiangiogenic, antiallergic, antiinflammatory and
43
antidiabetic compounds [Kim and Himaya, 2011; Zubia et al., 2008; Zhao et al., 2008; Li et
44
al., 2011; Gupta and Abu-Ghannam, 2011] and are a good candidates for use as functional
45
ingredients in the food industry.
46
The extraction of phlorotannins from brown seaweed and their characterization has
47
been reported previously [Ortiz et al., 2006; Sanchez-Machado et al., 2004; Glombitza and
48
Pauli, 2003; Tello-Ireland et al., 2011; Li et al., 2006; Gupta et al., 2011; Koivikko et al.,
49
2007; Leyton et al., 2016], although their purification from crude extracts has not been
50
widely studied yet. Several techniques have been used to purify phlorotannins, such as
51
chromatography using a Sephadex LH-20 column [Kantz and Singleton, 1990; Nwosu et
52
al., 2011; Koivikko et al., 2007], ultrafiltration using membranes with cut-off between 100
53
and 5 KDa [Wang et al., 2012] and liquid–liquid fractionation with ethyl acetate [Cho et al.,
54
2012; Kang et al., 2012; Shibata et al., 2004]. The main drawbacks of these techniques are
55
the use of non food-grade solvents, such as acetone, methanol and ethyl acetate, the low
56
selectivity due to the presence of other co-extracted compounds, the precipitation of
57
polyphenol–protein complexes [Siebert, 1999] and the high cost of these processes [Wang
58
et al., 2012].
59
An alternative that overcomes most of the limitations of these techniques is
60
macroporous resin separation [Kim et al., 2014]. In this method, polyphenols in aqueous
61
solutions are adsorbed on the resins due to hydrophobic binding and aromatic stacking. The
62
adsorbed polyphenols are later desorbed using a mixture of water with an organic solvent,
63
such as ethanol (food-grade); sugars present in the crude extract do not interact with the
64
resins, hence, they are easily removed with water. In addition, these resins are approved by
65
the Food and Drug Administration to produce food products [Scordino et al., 2003].
66
To design the process adequately, the mechanisms of adsorption and desorption of
67
the macroporous resin should be understood. Therefore, adsorption isotherms must be
68
determined and fitted using isotherm models such as Langmuir or Freundlich, from where
69
several thermodynamic parameters, such as isosteric heat, entropy and free energy, can be
70
derived [Allen et al., 2004; Sohn and Kim, 2005]. In addition, kinetic data can be fitted to
71
pseudo-first-order and pseudo-second-order kinetic to determine the adsorption efficiency
72
[Ho and Mckay, 1998].
73
Therefore, the aim of this work was to determine the type of macroporous resin and
74
the operating conditions that improve the purification of phlorotannins extracted from
75
brown seaweed Macrocystis pyrifera. Adsorption kinetics, static adsorption and desorption,
76
the adsorption isotherm and thermodynamic adsorption parameters of phlorotannins will be
77
evaluated.
78
2. Materials and methods
Preparation of phlorotannin extract 79
The brown seaweed employed was Macrocystis pyrifera collected from Chiloe, 30
80
km southeast of Puerto Montt, Chile, in June 2013, donated by Prof. A. Buschmann from
81
the Universidad de Los Lagos. The alga was dried at 40°C and ground size < 0.5 mm prior
82
to extraction. The extraction condition was optimized in our previous work: solution 0.5 M
83
of NaOH, solid/liquid ratio of 1/20 at 100°C for 3 h. The
84
Whatmann N°1 paper and the liquid phase stored at 4°C until use. The phlorotannin
85
concentration present in the extract was of 1,800 mg phloroglucinol equivalent (PGE)/L
86
[Leyton et al., 2017].
mixer was filtered using
87
Adsorbents 88
Six macroporous resins were tested to choose the one with the best
89
adsorption/desorption behavior: Diaion HP-20, Sepabeads SP-850, Amberlite XAD-7,
90
XAD-16N,
91
characteristics are shown in Table 1. All resins were washed with ethanol 70% at 25 °C for
92
12 h, priori to use.
93
XAD-4
and
XAD-2
(from
Sigma-Aldrich).
Their
physicochemical
Adsorption kinetics of phlorotannins 94
The adsorption kinetics determination was carried out as follows: 0.2 g of each resin
95
was mixed with 30 mL of phlorotannin extract under continuous agitation (300 rpm) at 25
96
°C, 1 mL of solution was removed at different time intervals to determine the concentration
97
of phlorotannins in the aqueous solution. The adsorption kinetics allow us to determine the
98
time of adsorption equilibrium. The adsorption capacity of phlorotannins on the resins at a
99
given contact time was calculated with the following equation:
100 = − ×
. 1
101 102
Where qt is the adsorption capacity at the given contact time t (mg PGE/g dry resin),
103
Ct is the concentration of phlorotannins in the solution at the given contact time (mg
104
PGE/mL), C0 is the initial concentrations (mg PGE/mL) of phlorotannins in the solution, Vi
105
is the volume of the initial sample solution (mL), and W is the resin weight (g).
106 107
To determine the adsorption efficiency of different resins, the kinetic models of
108
pseudo-first-order (Eq 2) and pseudo-second-order (Eq 3) were adopted [Ho and McKay,
109
1998; Simonin, 2016; Ho, 2016]:
110 111
Pseudo first order:
112
− = . + . 2
113 114 115
Pseudo second order:
= +
. 3
116 117
Where K1 and K2 are the constant rate of the pseudo-first-order-model and pseudo-
118
second-order-model, respectively; qe is the adsorption capacity at the adsorption
119
equilibrium (mg PGE/g dry resin) and t is time.
120
Adsorption of phlorotannins on the resins
121
In order to determine the best macroporous resin for the adsorption of
122
phlorotannins, a static adsorption was performed, where 2 g of each resin was put into a
123
tube with 30 mL of phlorotannin extract. The tube was shaken using a shaking incubator, at
124
300 rpm, at 25°C to reach adsorption equilibrium. After adsorption, the resins were filtered
125
for the subsequent desorption of phlorotannins, and the concentration of phlorotannins in
126
the extract was measured. The adsorption capacity was calculated according to the
127
following equation:
128 =
− . 4
129 130
Where qe is the adsorption capacity at the adsorption equilibrium (mg PGE/g dry
131
resin), and Ce is the equilibrium concentration (mg PGE/mL) of phlorotannins in the
132
solution. C0, Vi and W are the same as defined above.
133
Desorption of phlorotannins from the resins 134
In order to determine the best macroporous resin for desorption of phlorotannins, a
135
static desorption was performed. A first step was to determine the best ethanol
136
concentration for desorption of phlorotannins, and different concentrations of ethanol (10,
137
20, 40, 60, 80, 90 and 100% v/v) were screened in XAD-7HP. Once the best ethanol
138
concentration was determined, 30 mL of solvent was added to the different resins, and
139
shaken at 300 rpm and 25°C to reach desorption equilibrium. After desorption, the resins
140
were filtered and the concentration of phlorotannins in the extract was measured.
141
Desorption ratio, capacity and purification level were calculated according to the following
142
equations:
143 144
Desorption ratio "=
# # ∙ 100 . 5 −
145 146
Desorption capacity
# = # ∙
# . 6
147 148
Purification level (= ∗ 100 . 7 #
149 150
Where D is the desorption ratio (%), Cd is the concentration of phlorotannins in the
151
desorption solution (mg PGE/mL), and Vd is the volume of the desorption solution (mL).
152
C0, Ce and W are the same as defined above, qd is the desorption capacity of phlorotannins
153
(mg PGE/g) and P is the level of purification.
154
Adsorption isotherms of phlorotannins 155
In order to determine the optimum temperature for the adsorption of phlorotannins,
156
three adsorption isotherms were determined, for which 30 mL of extract of phlorotannins at
157
different initial concentrations were mixed with 2 g of resin and shaken at 300 rpm at
158
different temperatures, specifically 25, 35 and 45°C, to reach adsorption equilibrium. After
159
the adsorption, the concentration of phlorotannins in the extract was measured.
160 161
In order to select a suitable model for describing the adsorption properties of the different resins, the Langmuir and Freundlich equations were tested:
162 163
Langmuir: =
+, . 8 1 + +,
164 165
Freundlich:
= +/ 0 . 9 166 167
Where qm is the maximum adsorption capacity of the adsorbent (mg/g dry resin), KL
168
is the parameter related to the adsorption energy (L/mg), KF reflects the adsorption capacity
169
of an adsorbent (mg/g·(L/mg)1/n), and the parameter n represents the affinity of the
170
adsorbent for a given adsorbate.
Thermodynamic parameters 171
In order to determine the thermodynamic behavior of the adsorption on the resins,
172
changes in thermodynamic parameters, such as enthalpy (∆H), entropy (∆S), and free
173
energy (∆G) were determined. The change of ∆H and ∆S can be obtained from the slope
174
and intercept of the plot of the natural logarithm of the constant of adsorption equilibrium
175
(Keq) and 1/absolute temperature (1/T). ∆G was determined using the following equations:
176 ∆3 = −4567+ . 10 177 67+ = −
∆3 ∆8 ∆9 = − + . 11 45 45 4
178 179
Where R is the gas constant (J/mol K) and T is absolute temperature (°K).
180
Determin ing the phlorotannin content 181
The concentration of phlorotannins in the extracts was determined according to the
182
Folin–Ciocalteu assay [Singleton and Rossi 1965] adapted to 96-well plates, as well as
183
standards containing phloroglucinol with concentrations varying from 20 to 100 mg/L.
184
Samples and standards (20 µl) were introduced separately into 96-well plates, each
185
containing 100 µl of Folin–Ciocalteu’s reagent diluted with water (10 times) and 80 µl of
186
sodium carbonate (7.5% w/v). The plates were mixed and incubated at 45°C for 15 min.
187
The absorbance was measured at 765 nm using a UV–Visible spectrophotometer. The
188
phlorotannin concentration was determined by the regression equation of the calibration
189
curve and expressed as mg of phloroglucinol equivalent (mg PGE/L).
190 191
Determination of radical scavenging activity, total antioxidant activity
192
The free radical scavenging activity was measured using the modified method of
193
von Gadow et al., [1997]. 40 µl of 0.4 M 1,1-diphenyl-2-picryl-hydrazyl (DPPH) solution
194
in ethanol was added to 50 µl of the sample solution, supplemented with 110 µl of ethanol.
195
The plates were mixed and allowed to stand for 30 min in the absence of UV light to avoid
196
decomposition. The absorbance was measured at 520 nm against an ethanol blank.
197
Calibration curves of Trolox (0–24 mg/L) were prepared and the results were expressed as
198
the number of equivalents of Trolox (mg TE/L).
199 200
High Precision Liquid Chromatography (HPLC) detection of carbohydrate and
201
phloroglucinol
202
The quantification of phloroglucinol was carried out using a reverse phase C18
203
column (150 x 4.6 mm, 5 µm) and an HPLC system (Shimadzu, Model LC-10A) with
204
ultraviolet (UV) detector (operating at 280 nm). 20 µL of extract was analyzed using as
205
mobile phase 1% v/v formic acid in deionized water (solvent A) and acetonitrile (solvent
206
B), fed at a flow rate and temperature of 1 mL/min and 25°C according to the following
207
elution gradient: 0-15 min, 5% B; 15-75 min, 5-100% B; 75-85 min, 100% B and 85-90
208
min, 100-5% B [Sundberg et al., 2003]. Phloroglucinol (Sigma) was used as standard.
209
The quantification of carbohydrate was carried out using a HPX-87H column and a
210
Refractive Index Detector (RID) in HPLC. 10 µL of extract was analyzed using as mobile
211
phase of sulfuric acid 50 mM, fed at a flow rate and temperature of 0.5 mL/min and 55°C
212
with lineal elution. Glucose, mannitol, galactose, fucoidan, fucose and laminarin were used
213
as standard.
214 215
Statistical analysis
216
Each experiment was carried out in triplicate. The measurements were presented as
217
average ± standard deviation. The significance and relative influence of each resin in the
218
static adsorption and desorption of phlorotannins were determined using the analysis of
219
variance (ANOVA). The significance of the factors was determined at a 5% confidence
220
level.
221 222
3. Results
The adsorption kinetics of phlorotannins 223
The curves of the adsorption kinetics for each resin are presented in Figure 1. In this
224
figure a gradual increase in the adsorption of phlorotannins with time was observed, where
225
all resins reached adsorption equilibrium after approximately 8 to 9 h of contact time.
226
Kinetic data was fitted to pseudo-first-order and pseudo-second-order models as shown in
227
Table 1. According to the correlation coefficient, the pseudo-first-order model best
228
describes the adsorption kinetics of phlorotannins (Table 1).
229 230
The phlorotannin adsorption isotherm
231
The phlorotannin adsorption isotherms for different resins at 25, 35 and 45°C are
232
shown in Figure 2. The phlorotannin data for adsorption was fitted to the Langmuir and
233
Freundlich isotherm equations; the equation constants and correlation coefficients obtained
234
for each models are listed in Table 2. According to the correlation coefficient obtained from
235
each model, the model that better described the adsorption properties was the Freundlich
236
model; the correlation coefficients obtained were 0.80 for XAD-16N, 0.90 for HP-20 and
237
0.91 for SP-850. The best temperature for adsorption of phlorotannins by different resins
238
was 25°C. Static adsorption and desorption of phlorotannins
239
The best time and temperature for phlorotannin adsorption were 8 hours and 25°C,
240
respectively. Static adsorption and desorption under these conditions were carried out. The
241
adsorption and desorption capacity of phlorotannins obtained for the resins tested are
242
shown in Figure 3. The highest adsorption capacity was obtained with HP-20 followed
243
closely by XAD-16N and SP-850 with 190 ± 10, 183 ± 18 and 183 ± 31 mg PGE/g,
244
respectively. The lowest adsorption capacity was obtained with XAD-2, (156 ± 27 mg
245
PGE/g). The adsorption efficiency obtained for each polymer was similar, showing values
246
of 70.6±1.1, 69.5±2.0, 70.3±0.2, 69.7±0.1, 70.6±0.9 and 69.1±3.3% for HP-20, XAD-16N,
247
SP-850, XAD-2, XAD-4 and XAD-7, respectively.
248 249
The best desorption of phlorotannins was achieved (data not shown) with ethanol at
250
90% v/v; hence, this concentration was employed for the desorption experiments. The
251
desorption ratios obtained for HP-20, XAD-16N and SP-850 were 32.8 ± 2.1, 38.2 ± 7.7
252
and 27.7 ± 4.6%, respectively, with a desorption capacity of 47.3 ± 0.2, 77 ± 13 and 53 ±
253
14 mg PGE/g, respectively. The purification levels (recovery of phlorotannins) reached
254
with the resins were 24.9, 42.0, 29.0, 36.0, 32.4 and 32.4% for HP-20, XAD-16N, SP-850,
255
XAD-2, XAD-4 and XAD-7, respectively. The statistical analysis (ANOVA) showed that
256
the resins do not present a significant effect (p