Tải bản đầy đủ

Optimized methodology for alkaline and enzyme assisted extraction of protein from sacha inchi copy

Journal of Food Process Engineering ISSN 1745–4530

OPTIMIZED METHODOLOGY FOR ALKALINE AND ENZYMEASSISTED EXTRACTION OF PROTEIN FROM SACHA INCHI
(PLUKENETIA VOLUBILIS) KERNEL CAKE
ROSANA CHIRINOS1, MARTIN AQUINO1, ROMINA PEDRESCHI2 and DAVID CAMPOS1,3
1

Instituto de Biotecnologıa, Universidad Nacional Agraria La Molina-UNALM, Lima, Peru
lica de Valparaıso, Quillota, Chile
School of Agronomy, Pontificia Universidad Cato

2

3
Corresponding author.
TEL: 1051 1 6147800 ext. 436;
FAX: 1051 1 3495764;
EMAIL: dcampos@lamolina.edu.pe

Received for Publication February 9, 2016
Accepted for Publication April 21, 2016

doi:10.1111/jfpe.12412

ABSTRACT
The residue after oil extraction from sacha inchi (SI) presents a high protein
content of $59% that can be further exploited to extract proteins. In this study,
the protein extraction parameters for defatted SI cake meal (DSICM) were
optimized using alkaline and enzyme-assisted extractions. A central composed
design (CCD) was used to optimize the protein yield for both methods. The
obtained response surface models (RSM) produced a satisfactory fitting of the
results for both extraction methods (R2 5 0.9609–0.9761). For the alkaline
extraction method, the optimal SI protein extraction conditions corresponded to
54.2C, solvent/meal 42/1 (v/w) ratio, NaCl concentration of 1.65 M, pH 9.5 for 30
min and yielded 29.7% protein. For the enzyme-assisted method, optimal
extraction conditions corresponded to an enzyme concentration of 5.6%, 40.4 min
extraction, solvent/meal 50/1 (v/w) ratio, pH 9.0 and 50C and yielded 44.7%
protein and hydrolysis degree of 7.8%.

PRACTICAL APPLICATIONS
The cake obtained after oil extraction from SI seed is an important source of
protein. Thus, efforts should focus on the development of protein extraction
processes from the cake to add value to this by-product. Up to date, studies are
very limited. Results obtained in this study allowed the optimization of the protein
extraction process from SI cake meal. The enzyme-assisted protein extraction
resulted in a higher quantity of protein recovery ($1.5 fold more) than the alkaline
protein extraction. The optimized protein extraction process will allow the food
industry to obtain isolates or protein concentrates from SI cake meal to be used as
techno-functional, nutritional and/or functional agent.

INTRODUCTION
Proteins are macronutrients necessary for human beings and
they constitute an important nutritional contribution not
only as energy source but as source of nitrogen and essential
aminoacids. Proteins are also important because they confer
physicochemical, functional and organoleptic properties to
foods (Scopes 1986).
Nonconventional sources of protein (e.g., by products
from agroindustry) could render added value as functional
ingredients, for nutritional purposes to fortify foods and for
pharmaceutical and cosmetics applications. Currently, the
C 2016 Wiley Periodicals, Inc.


Journal of Food Process Engineering 00 (2016) 00–00 V

food industry is in need of alternative protein sources that
can compete with the actual protein sources that dominate
the market (Pszczola 2004). Within this context, sacha inchi
(SI) or Plukenetia volubilis is a highly oil-containing seed
(54%) and with a relatively high protein content (27%)
(Hamaker et al. 1992). The cake remaining after oil extraction from SI presents a protein content of 59% dry weight
(DW) (Sathe et al. 2012; Ruiz et al. 2013). Hamaker et al.
(1992) have reported in SI protein a high content of cysteine,
tyrosine, threonine and tryptophan and a low content of
phenylalanine. In addition, Sathe et al. (2012) reported that
1


OPTIMIZATION OF PROTEIN EXTRACTION FROM SACHA INCHI CAKE

the water soluble albumin fraction constituted $25% (%) of
the defatted SI seed flour.
In the last few years, there is an increasing interest on
methods for extracting plant protein based on acid, alkaline
and enzyme-assisted extraction (Sari et al. 2013), being the
alkaline and enzyme-assisted methods more amenable for
practical applications. The number of studies dedicated to
protein extraction from SI is very limited up to date and
none of the previous studies have focused on the optimization of the protein extraction parameters using the response
surface methodology (RSM). Sathe et al. (2012) evaluated
the defatted flour protein solubility of SI using a step by step
alkaline extraction with yields of $50% protein.
RSM is an excellent statistical technique for the optimization of complex processes (Box and Draper 2007). RSM
explores the existing relationships between explicative variables and one or more response variables (Cao et al. 2012).
This methodology has been previously used in the optimization of protein extraction either using alkaline or enzymaticassisted methods from different food sources such as flaxseed
(Oomah et al. 1994), pine seed (Wang et al. 2011), palm kernel cake (Chee et al. 2012), soybean (Rosenthal et al. 2001;
Rosset et al. 2014), lentil (Jarpa-Parra et al. 2014), etc.
Since there are no previous studies on the optimization of
protein extraction yields from SI under alkaline and enzymeassisted methods in combination with RSM, the objectives
of this study were (1) to evaluate the effect of alkaline extraction parameters such as NaCl concentration, temperature
and solvent:meal ratio at pH 9.5 on the response variable
protein yield (%) from defatted SI cake meal (DSICM) by
applying RSM; (2) to evaluate the effect of enzyme-assisted
extraction parameters with Alcalase 2.4L enzyme concentration and time at pH 9.0, on the response variables protein
yield and hydrolysis degree (%, HD) by applying RSM. The
optimization of the protein extraction parameters in DSICM
offers an alternative process for obtaining protein from a
nonconventional source (SI cake) and simultaneously this
agro-industrial by-product could be re-valorized.

R. CHIRINOS ET AL.

40C for 2 h in an oven then was packed in polyethylene bags
and stored at 4C until use.

Enzyme and Chemicals
Alcalase 2.4L was provided by Novozyme (Bagsvaerd, Denmark). All chemicals used were of reagent grade and purchased from Sigma (St Louis, MO) and Merck (Darmstadt,
Germany).

Protein Analyses
The soluble proteins were determined according to Lowry
et al. (1951) and total protein with the Kjeldhal method
(AOAC 1995). Protein yield (Y, %) was calculated as g of
soluble protein from extract/100 g of total protein of
DSICM.

Hydrolysis Degree
HD (%) was determined in each hydrolyzed sample using
the method of Adler-Nissen (1979) by assaying free amino
groups with 2,4,6-trinitrobenzenesulphonic acid (TNBS)
and using the following equation:
Â
Ã
HD ð%Þ 5 ðh=htot Þ3100 5 100 3 ðAN2 – AN1 Þ=Npb ;
where h is the number of peptide bonds broken, htot is total
number of bonds per unit weight, AN1 is the amino nitrogen
content of the protein substrate before hydrolysis (mg/g
protein), AN2 is the amino nitrogen content of the protein
substrate after hydrolysis (mg/g protein) and Npb is the
amino nitrogen content of the peptide bonds in the protein
substrate (mg/g protein) as determined after total hydrolysis
with 6 M HCl at 110C for 24 h. The values of AN2 and AN1
were obtained from a standard curve at 340 nm absorbance
versus mg/L amino nitrogen generated with L-leucine.

Protein Extraction

MATERIALS AND METHODS
Defatted Sacha Inchi Cake
SI kernel cake was provided by Olivos del Sur enterprise
(Lima, Peru). SI cake was obtained after oil extraction from
SI seed using an expeller. Proximate analysis was performed
in SI kernel cake according to the method of AOAC (1995)
for nuts and nut products. Protein content was calculated
using a conversion factor of 5.70 (Sathe et al. 2012). The
cake was ground in a hammer mill to obtain particles of
$500 lm. Ground cake meal was defatted for 12 h using
petroleum ether at a solvent/meal ratio of 10/1 (w/v) under
300 rpm stirring conditions. The DSICM was air-dried at
2

Alkaline Extraction of Sacha Inchi Protein. Protein
from DSICM was extracted using selected combinations of
independent variables: temperature (C), solvent/meal ratio
(v/w) and NaCl concentration (M) according to the experimental design. All protein extractions were performed at pH
9.5, 300 rpm agitation for 30 min. These parameters were
kept constant based on preliminary studies. After extraction,
solutions were immediately centrifuged at 4,000 3 g for 30
min at 4C. The supernatant were filtered through Whatman
filter paper No. 1 and the soluble protein and protein yield
(%) were quantified. All the experiments were carried out in
triplicate.
C 2016 Wiley Periodicals, Inc.
Journal of Food Process Engineering 00 (2016) 00–00 V


R. CHIRINOS ET AL.

OPTIMIZATION OF PROTEIN EXTRACTION FROM SACHA INCHI CAKE

TABLE 1. CENTRAL COMPOSITE DESIGN ARRANGEMENT AND EXPERIMENTAL AND PREDICTED PROTEIN YIELD VALUES FOR ALKALINE
EXTRACTION
Coded variables

Uncoded variables

Protein yield (Y) %

Run

x1

x2

x3

X1

X2

X3

Experimental

Predicted

1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19

21
1
21
1
21
1
21
1
21.68
1.68
0
0
0
0
0
0
0
0
0

21
21
1
1
21
21
1
1
0
0
21.68
1.68
0
0
0
0
0
0
0

21
21
21
21
1
1
1
1
0
0
0
0
21.68
1.68
0
0
0
0
0

40
70
40
70
40
70
40
70
29.8
80.2
55
55
55
55
55
55
55
55
55

20
20
50
50
20
20
50
50
35
35
9.8
60.2
35
35
35
35
35
35
35

0.5
0.5
0.5
0.5
2
2
2
2
1.25
1.25
1.25
1.25
0
2.51
1.25
1.25
1.25
1.25
1.25

13.06
11.59
16.53
23.95
22.92
20.52
26.58
23.02
22.61
23.67
13.85
27.5
13.43
26.87
28.56
27.51
28.08
27.93
27.83

11.38
12.68
18.45
23.61
24.17
19.52
26.4
25.62
22.27
22.71
14.49
25.56
13.28
25.72
28.03
28.03
28.03
28.03
28.03

X1, temperature; X2, solvent/meal ratio; X3, NaCl concentration.

Enzyme-Assisted Extraction of Sacha Inchi
Protein. Protein from DSICM was extracted using Alcalase
2.4L and the selected combinations of independent variables:
enzyme concentration (% enzyme in relation to the DSICM
protein content) and time (min) according to the experimental design. Protein extraction was carried at pH 9.0, 50C,
300 rpm stirring and at a solvent/meal ratio of 50/1 (v/w).
These parameters were kept constant as recommended for
Alcalase (pH and temperature) and the solvent/meal ratio
based on preliminary studies. After extraction, solutions
were immediately centrifuged at 4,000 3 g for 30 min at 4C.
The supernatants were filtered through Whatman filter
paper No. 1 and the soluble protein, protein yield (%) and
HD (%) were determined. All the experiments were carried
out in triplicate.

(Table 1). The conversion of real values to coded values was
as follows:
xi 5 ðXi – Xo Þ=DXi ;

where xi is the dimensionless value of an independent variable, Xi is the real value of an independent variable, Xo is the
real value of an independent value at the center point and
DXi is the step change.
A central composite design (CCD) was used to allow fitting a second-order model (Nakai et al. 2006). A total of 19
randomized runs that included five central points were performed (Table 1). The proposed model for the response variable (Y (%), protein yield) corresponded to:
y5b0 1

4
X
i51

Experimental Design and Statistical Analysis
Alkaline Extraction of Protein. The RSM was used to
determine the influence of three independent variables and
the optimal conditions for protein extraction from DSICM.
The effect of the variables temperature (X1), solvent/meal
ratio (X2) and NaCl concentration (X3) on the protein
extraction yield (dependent variable, Y %) was investigated.
The selection ranges within which each factor varied was
based on preliminary experiments (data not shown). Each
variable was coded at five levels: 21.68, 21, 0, 1 and 1.68
C 2016 Wiley Periodicals, Inc.
Journal of Food Process Engineering 00 (2016) 00–00 V

(1)

bi zi 1

4
X
i51

bii z2i 1

4
X

bij zi zj ;

(2)

i6¼j51

where b0 is the value of the adjusted response to the central
point of the design, bi , bii and bij are the linear, quadratic
coefficients and the intercept, respectively.
The optimum protein extraction conditions consisted on
determining the maximum protein extraction yield (maxima
desirability) through a combination of different variables or
factors. Predicted values (Y) were transformed into a desirability value (d). The generated RSM to obtain maximum
protein yield from DSICM was experimentally validated
with three experimental replicates and the obtained values
compared to the ones predicted by the RSM model. The
3


OPTIMIZATION OF PROTEIN EXTRACTION FROM SACHA INCHI CAKE

R. CHIRINOS ET AL.

TABLE 2. CENTRAL COMPOSITE DESIGN ARRANGEMENT AND EXPERIMENTAL AND PREDICTED PROTEIN YIELD AND DEGREE HYDROLYSIS
VALUES FOR ENZYME-ASSISTED EXTRACTION
Coded variables

Uncoded variables

Protein yield (Y) %

Hydrolysis degree (HD, %)

Run

x1

x2

X1

X2

Experimental

Predicted

Experimental

Predicted

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

21
1
21
1
21.41
1.41
0
0
0
0
0
0
0

21
21
1
1
0
0
21.41
1.41
0
0
0
0
0

2.00
5.00
2.00
5.00
1.38
5.62
3.50
3.50
3.50
3.50
3.50
3.50
3.50

15.00
15.00
45.00
45.00
30.00
30.00
8.79
51.21
30.00
30.00
30.00
30.00
30.00

28.97
34.45
29.18
40.33
28.79
42.58
28.41
31.85
34.80
35.00
34.91
35.07
35.18

28.60
34.80
28.51
40.37
29.46
42.23
28.36
32.23
34.99
34.99
34.99
34.99
34.99

0.96
4.03
3.75
5.38
1.56
7.72
2.97
6.44
4.07
4.25
4.49
4.03
3.72

0.92
5.01
3.91
6.55
1.71
6.42
2.54
5.74
4.12
4.12
4.12
4.12
4.12

X1, enzyme concentration (%); X2, time (min).

surface plots were generated by varying two variables within
the experimental range and holding the other constant
(zero) at the central point. All the statistical analysis were
carried out with Statgraphics Centurion XV software 15.2.06
(Stat Point Inc., VA).
Enzyme-Assisted Extraction of Protein. The RSM
was used to determine the influence of two independent variables on the optimal conditions for enzyme-assisted protein
extraction from DSICM. In addition, the influence of the
same variables on the HD (%) of protein was examined. The
effect of the variables: enzyme concentration (X1) and time
(X2) on the protein extraction yield (maximum) and protein
HD (minimum) were investigated. The selection ranges
within which each factor was varied based on preliminary
experiments (data not shown). Each variable was coded at
five levels: 21.41, 21, 0, 1 and 1.41 (Table 2). The conversion of real values to coded values was conducted as
described in Eq. (1) for the two evaluated responses (protein
yield and HD).
A central composite design (CCD) allowed fitting of a
second-order model. A total of 13 runs that included five
central points were performed (Table 2). The proposed
model for the response variables (second order polynomial),
desirability values, validation of RSM and generated surface
plots were calculated as described previously. A multiple
response optimization was performed to determine the combination of the experimental parameters (independent variables) that simultaneously maximize the protein yield and
minimize the protein HD. The obtained result was experimentally validated with three experimental replicates. The
optimization of two variables was displayed as an overlaid
contour plot. All the statistical analysis were carried out
4

with Statgraphics Centurion XV software 15.2.06 (Stat
Point Inc., VA).

RESULTS AND DISCUSSION
Proximal Composition and Protein Analysis
SI kernel cake presented 7.9% humidity and the contents of
protein, fat, fiber, ash and carbohydrates in dry weight
(DW) corresponded to 58.4, 8.9, 4.1, 5.7 and 22.7%, respectively, these values are close to the ones reported by Ruiz
et al. (2013). DSICM reached values of 61.9% of protein
(DW), this value was superior to the protein content
reported for other defatted meals obtained from soybean,
palm, and sesame (50, 16.8 and 42%, respectively) (Onsaard
et al. 2010; Chee et al. 2012; Rosset et al. 2014). Moure et al.
(2006) reported that protein content of defatted meals from
dehulled oilseeds depend on the seed type and ranges
between 35 and 60% (DW).

Optimization of Alkaline Extraction
The experimental design of five-levels, three-variable CCD
and the experimental results of protein extraction are shown
in Table 1. Protein yield varied from 11.5 to 28.5% (or from
7.1 to 17.6 g protein/100 g of DSICM). Using alkaline extraction and the RSM, protein recoveries from different defatted
cakes from oilseeds ranged between 10.9 and 32.6; 3.3 and
5.7; 12.3 and 16.5; and 40.8 and 58.7 g of protein/100 g for
flaxseed, pigeon pea, soybean and lentil (Oomah et al. 1994;
Jarpa-Parra et al. 2014; Tan et al. 2014).
The application of RSM yielded the following regression
equation, which is an empirical relationship between protein
yield (Y) and the evaluated variables (Eq. (3)):
C 2016 Wiley Periodicals, Inc.
Journal of Food Process Engineering 00 (2016) 00–00 V


R. CHIRINOS ET AL.

OPTIMIZATION OF PROTEIN EXTRACTION FROM SACHA INCHI CAKE

Y ð%Þ 5 241:894 1 0:9801 Ã X1 1 0:9972 Ã X2
1 29:3638 Ã X3 2 0:0086X1 2
1 0:0:0042X1 Ã X2 2 0:1323 Ã X1 Ã X3

(3)

2

2 0:0125 Ã X2 2 0:1074 Ã X2 Ã X3
2 5:35729 Ã X3 2 :
Analysis of variance (Table 1 and Supporting Information)
revealed that DCC application resulted in a highly significant
model (P < 0.000) indicative of a good generated response
model for optimization with R2 5 0.9609 and an adjusted
R2 5 0.9218. These coefficients suggest good fitting of the
model given that at least the R2 should be higher than
0.8000 (Joglekar and May 1999). Within the experimental
evaluated range, the factor time did not significantly affected
(P > 0.05) protein extraction yield meanwhile the other
components solvent/meal ratio and NaCl concentration had
a high significant effect (P < 0.01). These results indicate
that solvent/meal ratio and NaCl concentration are the main
factors contributing to protein extraction from DSICM.
Similar results have been previously obtained for extracted
protein from flaxseed and cowpea flour (Oomah et al. 1994;
Mune et al. 2008).
The surface responses are displayed in Fig. 1. The effect of
temperature and solvent/meal ratio on protein yield is displayed in Fig. 1a. Solvent/meal ratio exerted a quadratic
effect on protein yield that can be evidenced in Fig. 1a, where
its interaction with temperature is also displayed and with
solvent/meal ratios between 40/1 and 50/1 displaying the
highest protein yields. In Fig. 1a, a linear effect of temperature on protein yield can be observed, and within the 30–
70C range, no big variations were observed for the different
evaluated solvent/meal ratios. Temperature exerted a slight
quadratic effect on protein yield at different NaCl concentrations (Fig. 1b). The interaction of solvent/meal ratio and
NaCl concentration is displayed in Fig. 1c, where the quadratic effect of both components is evidenced. The quadratic
effect of NaCl concentration was also evidenced in Fig. 1c,
reaching the maximum protein yield at NaCl concentrations
close to 1.5 M. Results indicate that solvent/meal ratio and
NaCl concentration are the main contributors to the protein
extraction from DSICM.
The desirability maxima function was used to obtain the
optimal extraction conditions. The dependent variable was
set to the maximum possible (d 5 1), the optimal conditions
corresponded to 54C, a solvent/meal ratio of 42/1 (v/w),
NaCl 1.65 M at a pH of 9.5 and 30 min extraction time,
obtaining a protein yield of 29.7% (18.4 g protein/100 g
DSICM). Higher extraction yields (40.9 and 47 g solubilized
protein/100 g defatted SI flour) were reported by Sathe et al.
(2012) for SI meal defatted with hexane, using a step by step
methodology, at conditions of 1M NaCl, 15–30 min extracC 2016 Wiley Periodicals, Inc.
Journal of Food Process Engineering 00 (2016) 00–00 V

FIG. 1. RESPONSE SURFACE PLOTS AND CONTOURS FOR THE
EFFECTS OF (a) TEMPERATURE VERSUS SOLVENT/MEAL RATIO, (b)
TEMPERATURE VERSUS NaCl CONCENTRATION AND (c) SOLVENT/
MEAL RATIO VERSUS NaCl CONCENTRATION ON PROTEIN YIELD FOR
ALKALINE METHOD OF PROTEIN EXTRACTION

tion time, two extraction steps, pH 9–12 and room temperature. The differences with that study can be attributed to the
solvent/meal ratio and differences of the raw materials. The
SI cake used in our study was exposed to a mechanical force
and friction generated in the expeller during the process to
obtain SI oil that could have affected the physicochemical
characteristics of the SI protein disfavoring its extraction.
Finally, the suitability of the generated mathematical model
to predict maximum protein yield was experimentally validated using the conditions determined in the optimization.
Thus, the experimental protein yield at the optimum conditions was 30.2 6 0.33% being this value close to the value
generated by the mathematical model.

Optimization of Enzyme-Assisted Extraction
This study evaluated Alcalase 2.4L with the aim to increase
the protein yield from DSICM with a low HD. A low HD is
key to obtain a not highly hydrolyzed protein that can be
used as raw material in different food applications.
5


OPTIMIZATION OF PROTEIN EXTRACTION FROM SACHA INCHI CAKE

The experimental design corresponded to five-levels and
two CCD variables. The experimental results of protein yield
as well as the HD are shown in Table 2. Protein yield and
HD varied from 28.4 to 40.3% (or from 17.3 to 24.5 g protein/100 g of DSICM) and from 0.96 to 5.38%, respectively.
Protein yield significantly increased ($40%) when the Alcalase was employed in the protein extraction process in comparison to the alkaline extraction. The efficiency of
proteolytic enzymes during protein extraction from different
sources has been extensively reported. Sari et al. (2013) by
using different proteases (1% of enzyme for 3 h) extracted
more protein from rapeseed, microalgae and soybean meals
($60, 80 and 90%, respectively) in comparison to alkaline
extraction (pH 9.5; $15, 30 and 80%, respectively). A significantly (P < 0.05) higher trypsin extracted protein yield
(61.9 g/100 g) was obtained from palm kernel in comparison
to the alkaline (pH 9.5) method (10.2 g/100 g) (Chee et al.
2012). Also Latif and Anwar (2011) found that proteases
Protex 7L and Alcalase 2.4L successfully extracted proteins
from sesame meal (87.1 and 79.6%, respectively). For the
HD, the maximum obtained corresponded to 5.38%. Sari
et al. (2013) reported that certain amount of hydrolysis is
needed and acceptable for protein extraction but a high
hydrolysis is detrimental because proteins would be converted to peptides displaying an increased solubility and
thus altering the functional properties of the extracted proteins (Rosenthal et al. 2001; Taha and Ibrahim 2002) and the
bitterness associated with a high HD. Taha and Ibrahim
(2002) reported that low protein HD (range between 8.8
and 9.5%) for soybean, sesame and rice bran meals enzymatically hydrolyzed with papain (0.06–0.21%) for 5 min
produced improvements in wettability, flow ability and
emulsifying capacity properties and a direct relation between
increasing HD, nitrogen solubility and dispersibility was
found.
A quadratic (Eq. (4)) and lineal (Eq. (5)) relationship was
found between protein yield and HD with the different
extraction parameters evaluated by SRM. The relationship
established between protein yield (Y) and HD (in real values) with the evaluated parameters is shown:
Y ð%Þ 5 21:2596 2 0:2097 Ã X1 1 0:4973
à X2 1 0:1901X1 2 1 0:063X1 à X2 – 0:0104 à X2 2 ;

(4)

HD ð%Þ 5 22:0681 1 1:1176 Ã X1 1 0:0753 Ã X2 : (5)
Analysis of variance (Table 2 and 3 in Supporting Information) revealed that CCD application resulted in a highly significant model (P < 0.000) indicative of a good generated
response model for optimization with a good R2 5 0.9934
and the adjusted R2 5 0.9887 for protein yield and with a
moderate R2 5 0.8551 and adjusted R2 5 0.8261 for HD.
Within the experimental evaluated range, the factors enzyme
6

R. CHIRINOS ET AL.

FIG. 2. RESPONSE SURFACE PLOTS AND CONTOURS FOR THE
EFFECTS OF ENZYME CONCENTRATION AND EXTRACTION TIME ON
(a) PROTEIN YIELD AND (b) HYDROLYSIS DEGREE FOR THE ENZYMEASSISTED METHOD OF PROTEIN EXTRACTION

concentration and time significantly affected (P < 0.05) protein extraction yield as well as HD. Thus enzyme concentration and time contribute to protein extraction from DSICM
and protein HD. Similar results were reported for soybean
meal, soybean, sesame and rice bran meals and palm kernel
meal (Rosenthal et al. 2001; Taha and Ibrahim 2002; Chee
et al. 2012).
The surface response for protein yield is displayed in
Fig. 2a. The effect of the enzyme concentration (%) in the
evaluated range on protein yield presented an increasing
trend (from $20 to 28%) as the enzyme concentration was
incremented (from 1.38 to 5.62%). Time exerts a quadratic
effect on protein yield. High protein yields were observed
between 40 and 50 min of extraction. The dependent variable
was set to the maximum possible (d 5 1.00) and the obtained
optimal conditions corresponded to a time of 54 min and
enzyme concentration of 5.5% considering a 50/1 (v/w) solvent/meal ratio, pH 9.0 and 50C, respectively, obtaining a
protein yield of 43.4% (26.8 g protein/100 g DSICM). Using
the predicted optimum conditions, experiments carried out
in triplicates gave good results (43.78 6 0.28%) that coincided with the predicted value implying that the model was
adequate. The surface response for HD is displayed in
Fig. 2b. The effect of the enzyme concentration (%) in the
evaluated range of HD presented a lineal effect. Increasing
concentrations of Alcalasa 2.4L resulted in higher HD. Also
time exerted a lineal effect but its effect was less pronounced
on HD. The optimization consisted on a minimization
(d 5 0) because a low as possible HD was aimed. The
obtained optimal conditions corresponded to 8.78 min and
C 2016 Wiley Periodicals, Inc.
Journal of Food Process Engineering 00 (2016) 00–00 V


R. CHIRINOS ET AL.

OPTIMIZATION OF PROTEIN EXTRACTION FROM SACHA INCHI CAKE

FIG. 3. SUPERIMPOSED CONTOUR PLOT FOR PROTEIN YIELD AND
HYDROLYSIS DEGREE (HD) AS A FUNCTION OF ENZYME
CONCENTRATION (%) AND EXTRACTION TIME (min) AT 50C,
SOLVENT/MEAL RATIO OF 50/1 AND pH 9.0

conditions, enzyme concentration of 5.6%, 40.4 min extraction, solvent/meal 50/1 (v/w) ratio, pH 9.0 and 50C) was
46% higher in comparison to the alkaline method (optimal
conditions, temperature: 54.2C, solvent/meal 42/1 (v/w)
ratio, NaCl concentration of 1.65 M, pH 9.5 for 30 min).
The predicted values for protein yield from all generated
models were consistent and experimentally validated. These
results indicate that the enzyme-assisted protein extraction
from sacha inchi kernel cake is an alternative protein extraction method with higher yields than the traditional alkaline
method. Additionally, the recovered protein from this byproduct could be considered as potential source of proteins
to be used in multiple industrial applications.

ACKNOWLEDGMENT
1.37% of enzyme concentration obtaining a HD of 0.13%.
Using the predicted optimum conditions, experiments carried out in triplicate gave good results (1.51 6 0.11%) very
close to the values predicted by the generated SRM model.
Finally, the obtained optimization results for both responses
did not offer concluding results when they were evaluated in
separate. Thus, a multiple optimization response was generated and the factors time and enzyme concentration that
resulted in a high protein yield and a low HD (lower to 10%)
were included. The optimization of these two responses is
displayed as an overlaid contour plot in Fig. 3. After the multiple response optimization, values of 5.62% Alcalase 2.4L
enzyme and 40.4 min at pH 9.0, 50C and 50/1 solvent/meal
ratio resulted in a maximum protein yield of 44.7% (27.6 g
protein/100 g DSICM) with a HD of 7.86%. Same conditions
were experimentally validated resulting in protein yield and
HD of 44.7 6 0.4 and 7.86 6 0.14%, respectively. Our results
indicate that the enzyme-assisted protein extraction was able
to extract 1.46 fold more protein than the alkaline extraction
from DSICM.

CONCLUSIONS
RSM allowed optimization of the alkaline and enzymeassisted protein extraction conditions from DSICM. For the
protein alkaline method, the factors: solvent/meal ratio and
NaCl concentration significantly affected the extraction conditions, but not extraction time. For the enzyme-assisted
protein extraction method, the Alcalase 2.4L enzyme concentration and time of hydrolyses affected the protein yield
and the HD. By means of a multiple response methodology
(MRM) with the responses: protein yield and HD which
were maximized and minimized, respectively, it was possible
to obtain the maximum protein extraction with a low HD.
Results of the MRM for the enzyme-assisted protein extraction method indicated that maximum protein yield (optimal
C 2016 Wiley Periodicals, Inc.
Journal of Food Process Engineering 00 (2016) 00–00 V

This research was supported by the grant in Science and
Technology (2013–2014) supported by the Universidad
Nacional Agraria La Molina (Lima, Peru).
REFERENCES
ADLER-NISSEN, J. 1979. Determination of the degree of hydrolysis of food protein hydrolysates by trinitrobenzenesulfonic acid.
J. Agric. Food Chem. 27, 1256–1262.
AOAC. 1995. Officials Methods of Analysis, 15th Ed., Association
of the Official Analytical Chemists, Washington, D.C., Gaithersburg, Maryland.
BOX, G. and DRAPER, M. 2007. Response Surfaces, Mixtures, and
Ridge Analyses. 2nd Ed., John Wiley & Sons Inc., Hoboken,
New Jersey.
CAO, W., ZHANG, C., JI, H. and HAO, J. 2012. Optimization of
peptic hydrolysis parameters for the production of angiotensin
I-converting enzyme inhibitory hydrolysate from Aceteschinensis through Plackett–Burman and response surface methodological approaches. J. Sci. Food Agric. 92, 42–48.
CHEE, L., LING, H.K. and AYOB, K. 2012. Optimization of
trypsin-assisted extraction, physochemical characterization,
nutritional qualities and functionalities of palm kernel cake
protein. LWT - Food Sci. Technol. 46, 419–427.
HAMAKER, B.R., VALLES, C., GILMAN, R., HARDMEIER,
R.M., CLARK, D., GARCIA, H., GONZALES, A.E.,
KOHLSTAD, I., CASTRO, M., VALDIVIA, R., et al., 1992.
Amino acid and fatty acid profiles of the inca peanut (Plukenetia volubilis). Cereal Chem. 69, 461–463.
JARPA-PARRA, M., BAMDAD, F., WANG, Y., TIAN, Z., TEMELI,
F., HAN, J. and CHEN, L. 2014. Optimization of lentil protein
extraction and the influence of process pH on protein structure
and functionality. LWT - Food Sci. Technol. 57, 461–469.
JOGLEKAR, M. and MAY, T. 1999. Product excellence through
experimental design. In Food Product and Development: From
Concept to the Market Place (E. Graf and I.S. Saguy, eds.), Aspen
Publishers Inc., Gaithersburg, Maryland.
7


OPTIMIZATION OF PROTEIN EXTRACTION FROM SACHA INCHI CAKE

LATIF, S. and ANWAR, F. 2011. Aqueous enzymatic sesame oil
and protein extraction. Food Chem. 125, 679–684.
LOWRY, H., ROSEBROUGH, J., FARR, L. and RANDALL, J.
1951. Protein measurement with the Folin phenol reagent. J.
Biol. Chem. 193, 265–275.
 J.
MOURE, A., SINEIRO, J., DOMINGUEZ, H. and PARAJO,
2006. Functionality of oilseed protein products: A review. Food
Res. Int. 39, 945–963.
MUNE, M.A., MINKA, S.R. and MBOME, I.L. 2008. Response
surface methodology for optimisation of protein concentrate
preparation from cowpea [Vigna unguiculata (L.)Walp]. Food
Chem. 110, 735–741.
NAKAI, S., LI-CHEN, Y. and DOU, J. 2006. Experimental design
and response surface methodology. In Handbook of Food and
Bioprocess Modeling Techniques (S. Sablani, A. Datta, M.S. Rahman and A. Mujumdar, eds.), CRC Press, Boca Raton, FL.
ONSAARD, E., POMSAMUD, P. and AUDTUM, O. 2010. Functional properties of sesame protein concentrate from sesame
meal. Asian J. Food Agro-Ind. 3, 420–431.
OOMAH, B.D., MAZZA, G. and CUI, W. 1994. Optimization
of protein extraction from flaxseed meal. Food Res. Int. 27,
355–361.
PSZCZOLA, D. 2004. Ingredients of food technology. J. Food Sci.
58, 56–69.
ROSENTHAL, A., PYLE, D.L., NIRANJAN, K., GILMOUR, S.
and TRINCA, L. 2001. Combined effect of operational variables and enzyme activity on aqueous enzymatic extraction of
oil and protein from soybean. Enzyme Microb. Technol. 28,
499–509.

8

R. CHIRINOS ET AL.

 P. 2014. Protein extracROSSET, M., ACQUARO, R. and BELEIA,
tion from defatted soybean flour with Viscozyme L pretreatment. J. Food Process. Preserv. 38, 784–790.
RUIZ, C., DIAZ, C., ANAYA, J. and ROJAS, R. 2013. Analisis
proximal, antinutrientes, perfil de acidos grasos y de aminoacidos de semillas y tortas de 2 especies de sacha inchi (Plukenetia
volubilis y Plukenetia huayllabambana). Rev. Sociedad Quımica
Per
u 79, 29–36.
~ IS, M. and SANDERS, J. 2013. Enzyme assisted
SARI, Y., BRUN
protein extraction from rapeseed, soybean and microalgae
meals. Ind. Crops Prod. 43, 78–83.
SATHE, S., KSHIRSAGAR, H. and SHARMA, G. 2012. Solubilization, fractionation, and electrophoretic characterization of Inca
Peanut (Plukenetia volubilis L.) proteins. Plant Foods Hum.
Nutr. 67, 247–255.
SCOPES, R. (1986). Protein Purification. Principles and Practice,
3rd Ed., Springer, New York.
TAHA, F.S. and IBRAHIM, M.A. 2002. Effect of degree of hydrolysis on the functional properties of some oilseed proteins.
Grasas y Aceites 53, 273–281.
TAN, E.S., NGOH, Y.Y. and GAN, C. H.Y. 2014. A comparative
study of physicochemical characteristics and functionalities of
pinto bean protein isolate (PBPI) against the soybean protein
isolate (SPI) after extraction optimization. Food Chem. 152,
447–455.
WANG, S., JIANG, L., LI, Y., LI, D. and SUI, X. 2011. Optimization on aqueous enzymatic extraction conditions of pine
seed protein by response surface method. Proc. Eng. 15,
4956–4996.

C 2016 Wiley Periodicals, Inc.
Journal of Food Process Engineering 00 (2016) 00–00 V



Tài liệu bạn tìm kiếm đã sẵn sàng tải về

Tải bản đầy đủ ngay

×