Solvent optimization for anthocyanin extraction from Syzygium cumini

Prévia do material em texto

Solvent optimization for anthocyanin extraction from Syzygium cumini
L. Skeels using response surface methodology
BRATATI CHAUDHARY & KUNAL MUKHOPADHYAY
Department of Biotechnology, Birla Institute of Technology, Mesra 835215, Ranchi, India
Abstract
Anthocyanins are plant pigments that are potential candidates for use as natural food colourant. In this study, Syzygium cumini
fruit skin has been used as anthocyanin source. All the six major types of anthocyanins were identified in the sample by ultra
performance liquid chromatography studies, and the antioxidant activity was found to be 4.34 ^ 0.26 Fe2þg21 in the sample
with highest anthocyanin content. Optimization of conditions for extracting high amounts of anthocyanin from the fruit peels
was investigated by response surface methodology. The results suggested that highest anthocyanin yield (763.80 mg; 100 ml21),
highest chroma and hue angle in the red colour range could be obtained when 20% ethanol was used in combination with 1%
acetic acid. Methanol was replaced with ethanol for the extraction of pigments due to its less toxicity and being safe for human
consumption. The optimized solvent can be used to extract anthocyanins from the S. cumini fruits and used as natural colourants
in the food industries.
Keywords: Plakett–Burman, quadratic, factorial, colourant, antioxidant, pigment
Introduction
Colours used in the food processing industries to make
the food more attractive and appealing (Noonan 1972)
are generally synthetic colourants that have raised legal
and health issues over the last few years. Consumers
now demand natural substitutes of the synthetic
colourants as more healthy and natural ingredients in
the foods they consume (Guisti and Wrolstad 2003).
Plant-derived natural colours from fruits and flowers
can be used to replace these synthetic colourants (Vyas
et al. 2009). Anthocyanins are water-soluble pigments
which provide attractive colours such as pink, red,
orange, blue and purple to some fruits and flowers
(Grotewold 2006). These pigments are natural
antioxidants with several health benefit properties
(Chalabi et al. 2008; Chaudhary and Mukhopadhyay
2012a). Thus, they are potential candidates for use as
natural colourants in the food processing industries
(Mortensen 2006). The colour of anthocyanin
extracts is pH dependent, and thus, the solvent plays
an important role in optimum extraction of these
pigments. Methanol with hydrochloric acid (HCL) is
commonly used to extract anthocyanins, but methanol
is not safe for human consumption (Huang et al.
2010). To make the extract edible, food grade ethanol,
which has a lower toxicity, can be used in combination
with low concentration of acids. Anthocyanins have
been extracted from deep purple berries of black
currant plants by using different combinations of
acidic organic solvents (Kapasakalidis et al. 2006).
The conventional optimization experiments are
conducted by selecting one factor and one response
at a time. This method is usually inefficient and yields
misleading results. By performing a factorial design,
the most significant set of experiments with minimal
number of runs can be designed, and the effect of
multiple factors on various responses can be studied
simultaneously by using response surface method-
ology (RSM) (Ahmed et al. 2011; Zou et al. 2011).
This method also allows identification of the inter-
actions in a process and helps to reduce the use of
ISSN 0963-7486 print/ISSN 1465-3478 online q 2012 Informa UK, Ltd.
DOI: 10.3109/09637486.2012.738647
Correspondence: Kunal Mukhopadhyay, Department of Biotechnology, Birla Institute of Technology, Mesra, Ranchi 835215, Jharkhand, India.
Tel: þ 91-9431382720. Fax: þ 91-651-2275401. E-mail: kmukhopadhyay@bitmesra.ac.in
International Journal of Food Sciences and Nutrition,
May 2013; 64(3): 363–371
hazardous chemicals, improves product quality and
reduces production cost. The theoretical aspects of
selecting an experiment model in RSM have been
reviewed, and it was found that the selection of
parameters and their levels plays an important role
(Bas and Boyaci 2007).
RSM has been widely used in chemical and
biological processes to predict the optimal conditions
of various extraction methods. Two types of models
can be applied in RSM depending on the
relationship between factors and response (Huang
et al. 2010). In case of a linear relation between the
independent factors and response, a first-order model
can be used:
Y ¼ b0 þ
Xk
i¼1
biXi þ 1; ð1Þ
where Y is the response, Xi is the input variable, b0 is
the intercept, bi is the linear coefficient and 1 is an
error term.
If the relationship is a curve, the second-order
model can be introduced to fit the system. The
second-order model includes linear, quadratic and
interactive components:
Y ¼ b0 þ
Xk
i¼1
biXi þ
Xk
i¼1
biiX
2
i
þ
Xk
i¼1
Xk
i,j
bijXiXj þ 1;
ð2Þ
where Y is the response, Xi and Xj are input variables,
b0 is the intercept, bi is the linear coefficient, bii is the
squared coefficient and 1 is an error term.
In this study, Syzygium cumini L. Skeels, family
Myrtaceae, has been used as a source of anthocyanin
pigment. S. cumini fruits have a purplish black
appearance and have high anthocyanin content. The
anthocyanin accumulates in the fruit skin as well as in
the flesh (Ayyanar and Babu 2012; Chaudhary and
Mukhopadhyay 2012b). Five types of anthocyanins,
i.e. delphinidin, cyanidin, malvidin, peonidin and
petunidin, have been identified in the fruit in their
diglucoside forms (Brito et al. 2007; Veigas et al.
2007). A Plakett–Burman design was used, and the
influence of independent factors on the anthocyanin
extraction process was analysed by using RSM with
the objective to minimize the use of quantities of acids
and solvents.
Methods
Plant material
Ripe fruits of S. cumini were collected from a tree
located in the campus of Birla Institute of Technology,
Mesra, Ranchi, India, and were used to carry out the
experiment. The peels of the ripe berries were
manually separated, weighed and flash frozen using
liquid nitrogen. The peels were ground into a fine
powder using sterile mortar and pestle.
Extraction and quantification of anthocyanin
Different sets of solvents were prepared for the
experiment, and anthocyanin was extracted from
powdered fruit peels. A set of 46 manual experiments
with different solvent combinations were designed.
Different solvent systems varied in the solvent
percentage (0, 20, 50, 80 and 100), acid type (HCl,
acetic acid and citric acid) and acid percentage (0.25,
0.5 and 1). One extraction with only MilliQ water was
also done which served as a control. Ethanol was
chosen as the solvent and MilliQ water was used to
make up the different solvent concentrations. Food
grade ethanol and acids used for the study were
procured commercially. Anthocyanins being polar
molecules are soluble in polar solvents such as water
and alcohol. In this study, the main objective was to
optimize a solvent system with minimal use of
chemicals. Water also is a good solvent for anthocya-
nins, but anthocyanin extracted in water loses its
stability and is therefore not used for food colouring
purposes. When ethanol is added to water, the extracts
have better quality characteristics (Delgado-Vargas
and Paredes-Lopez 2003).
Anthocyanins were extracted by soaking 5 g of fresh
fruit peel powder in 50 ml of various solvent sets for 4 h
on an orbital shaker set at 100 rpm at 48C to minimize
degradation of isolated anthocyanins. The same
amount of fruit peel was used for all the solvent sets.
After filtration, the residue was repeatedly extracted
with the same volume of solvent until the extract
obtained was almost colourless. The extract from a
single type of solvent was combined and purified using
Sep-Pak plus short tC18 cartridges (Waters Corpor-
ation, Milford, MA, USA). The volume of the purified
extracts was adjusted to 100 ml,and the extracts were
used to quantify total anthocyanin content using the
pH differential method (Lee et al. 2005) and were
expressed as mg 100 ml21 (Veigas et al. 2007).
Experimental design for optimization
RSM was used to optimize the solvent composition for
extracting anthocyanins from S. cumini fruit skin.
Design-Expert version: 8.0.7.1 (Stat-Ease Inc., Min-
neapolis, MN, USA) was used to design the
experiment. The solvent percentage, acid type and
acid percentage were the parameters that were
Table I. Coded variables of factors and their levels.
Variables Level (21) Level (0) Level (þ1)
Solvent (X1) (%) 20 50 80
Acid concentration (X2) (%) 0.25 0.50 1
Acid type (X3) HCl Acetic acid Citric acid
B. Chaudhary and K. Mukhopadhyay364
considered while designing the optimization exper-
iment model. The main objective of designing the
experiment was to optimize the best solvent system to
extract maximum amount of anthocyanins. A factorial
design with three factors (X1, X2 and X3) and three
levels coded as 21, 0 and þ 1 (Table I) was employed
to statistically investigate the interactive responses of
the factors and their respective levels with each other
and also with the responses. Three responses (Y), i.e.
total anthocyanin content, chroma and hue angle,
were selected to determine the best solvent system for
the extraction of anthocyanins (Yang et al. 2008).
A total of 22 experiments (Table II) with 5 replicates
at the centre point were designed, which allowed
the estimation of a pure error sum of squares.
The multiple regression equation was used to fit
the second-order polynomial equation based on the
experimental data. For three independent variables
analysis (X1, X2 and X3), Equation (2) can be
rearranged as follows:
Y ¼ b0 þ b1X1 þ b2X2 þ b3X3 þ b12X1X2
þ b23X2X3 þ b13X1X3 þ b11X
2
1
þ b11X
2
2 þ b11X
2
3;
ð3Þ
where Y is the response variable, b0 is the intercept, b1,
b2, b3, b11, b22, b33 and b12, b13, b23 are linear,
quadratic and interaction coefficients, respectively,
and X1, X2 and X3 are the coded independent
variables. The same software Design-Expert was used
to determine the analysis of variance (ANOVA) and
the coefficient of determination (R 2) to evaluate the
goodness of fit of the model. A set of 46 manual
experiments were conducted and evaluated by the
same three responses, i.e. total anthocyanin content,
chroma and hue angle.
Colour characteristics
All the extracted anthocyanin samples were checked
for their colour value in terms of a*, b* and L*. To
evaluate the different colour patterns of the anthocya-
nin extracts, the tristimulus colorimetry CIEL*a*b*
scale of the Commission Internationale d’Eclairage
(CIE 1986) was followed using a colorimeter (Color-
flex, HunterLab, Reston, VA, USA) under the
condition C (400–700 nm, 7400 K), and data were
obtained at 28 viewing angle and D65 illumination.
Chroma (C*) and hue angle (hab) were calculated
from a*, b* and L* values applying the formula
C* ¼ (a*
2 þ b*
2)0.5 and hue angle hab ¼ tan
21
(b*/a*). The 2D and 3D chromaticity plots were also
used to analyse the results. The instrument operation
and the various data analysis such as 2D plot, 3D plot
and scanning at 400–700 nm were done using
Table II. Experimental design layout as obtained through design
expert software.
Runs Solvent percentage Acid percentage Acid type
1 50.00 1.00 HCl
2 80.00 1.00 AA
3 20.00 0.50 HCl
4 50.00 0.50 AA
5 50.00 0.50 CA
6 80.00 1.00 CA
7 50.00 0.25 HCl
8 80.00 0.50 HCl
9 80.00 0.50 HCl
10 50.00 0.25 HCl
11 80.00 1.00 HCl
12 20.00 0.25 CA
13 20.00 0.25 AA
14 50.00 1.00 HCl
15 20.00 1.00 CA
16 50.00 0.50 AA
17 50.00 0.50 AA
18 50.00 0.50 CA
19 80.00 0.25 CA
20 20.00 0.50 HCl
21 80.00 0.25 AA
22 20.00 1.00 AA
Notes: HCL, hydrochloric acid; AA, acetic acid; CA, citric acid.
Figure 1. Anthocyanin extracts of S. cumini fruit peels obtained from different solvent systems: (a) T.S. of S. cumini fruit, anthocyanin extracts
of (b) MilliQ water, (c) 100% ethanol þ 0.1% HCl, (d) 20% ethanol þ 1% HCl, (e) 20% ethanol þ 1% citric acid and (f) 20% ethanol þ 1%
acetic acid.
Solvent optimization for anthocyanin extraction from Syzygium cumini L. 365
EasyMatch QC software. All the samples were
analysed in triplicates.
Ferric-reducing antioxidant power assay
The total antioxidant activity of the fruit peel
extracted with the solvent that showed the highest
concentrations of anthocyanin was calculated using
the ferric-reducing antioxidant power (FRAP) assay
(Benzie and Strain 1999). FRAP reagent [prepared
freshly by mixing 10 volumes of sodium acetate
buffer pH 3.6, 1 volume of 10 mM TPTZ {2,4,6-tri
(2-pyridyl)-s-triazine} (Sigma Life Sciences, St
Louis, MO, USA) prepared in 40 mM HCl and 1
volume of 20 mM FeCl3,6H2O] was used to
calculate the antioxidant power of the fruit peel.
Ascorbic acid was used as standard having an FRAP
value of 1. A standard curve for the absorbance
values of L-ascorbic acid was plotted at different
concentrations (100–1000 mm) with the same
method and conditions. All the experiments were
carried out in triplicates. The difference in the
increase in the absorbance of the samples with
respect to the standard was determined and used to
calculate the FRAP values that were expressed as
mmole Fe2þg21 fresh weight of the samples.
Ultra performance liquid chromatography analysis
Anthocyanins present in the sample were separated
using a Waters Acquity ultra performance liquid
Table III. ANOVA for the overall effect of factors on response.
Factors DF SS MS F-value p Value
Total anthocyanin content (Y1)
X1 (solvent %) 3 0.55 0.55 22.64 0.0001
X2 (acid %) 3 0.32 0.32 17.62 0.0004
X3 (acid type) 3 0.44 0.44 25.21 0.0002
Chroma (Y2)
X1 (solvent %) 3 0.89 0.89 10.56 0.0002
X2 (acid %) 3 0.77 0.77 11.21 0.0001
X3 (acid type) 3 0.82 0.82 8.52 0.0004
Hue angle (Y3)
X1 (solvent %) 3 0.45 0.45 14.22 0.0010
X2 (acid %) 3 0.52 0.52 21.50 0.0012
X3 (acid type) 3 0.55 0.55 19.81 0.0031
Notes: DF, degree of freedom; SS, sum of squares; MS, mean of
squares.
Normal plot of residuals(a) (d)
(b) (e)
(c) (f)
99
95
90
80
70
50
30
20
10
5
1
5.00
4.00
3.00
2.00
1.00
3.00
2.00
1.00
0.00
–1.00
–2.00
–3.00
1.00 2.00 3.00 4.00 5.00
1.50 2.00 2.50 3.00 3.50 4.00 4.50
N
or
m
al
 %
 p
ro
ba
bi
lit
y
P
re
di
ct
ed
In
te
rn
al
ly
 s
tu
de
nt
iz
ed
 r
es
id
ua
ls
–3.00 –2.00 –1.00 0.00
Predicted vs. actual
1.00 2.00
Actual
Residuals vs. perdicted
Predicted
22
2
2
3
22
2
2
3
1000.0
800.0
600.0
400.0
200.0
0.0
1.00
1000.0
800.0
600.0
400.0
200.0
0.0
1.00
0.81
0.63
0.44
0.25 20.00
26.00
32.00
38.00
44.00
50.00
56.00
62.00
68.00
74.00
80.00
20.00
26.00
32.00
38.00
44.00
50.00
56.00
62.00
68.00
74.00
80.00
20.00
26.00
32.00
38.00
44.00
50.00
56.00
62.00
68.00
74.00
80.00
A: Solvent
A: Solvent
A: Solvent
B: Acid conc
B: Acid conc
B: Acid conc
A
nt
ho
cy
an
in
 c
on
te
nt
400.0
300.0
200.0
100.0
0.0
1.00
0.81
0.63
0.44
0.25
0.81
0.63
0.44
0.25
–100.0
C
hr
om
a
H
ue
 a
ng
le
Figure 2. The point prediction plots and 3D graphs of the quadratic model where (a) is the normal plot of residuals, (b) is the predicted versus
actual values, (c) the residuals versus predicted values, (d) the 3D plot of the effect of acid concentration on anthocyanins content, (e) the
chroma values and (f) the hue angle.
B. Chaudhary and K. Mukhopadhyay366
chromatography (UPLC) system (Waters Corpor-
ation) consisting of an Acquity UPLC binary systems
manager, an Acquity UPLC sample manager, an
Acquity Tunable UV (TUV) detector and an H/T
column heater containing an Acquity UPLC BEH C18
reverse phase column (2.1 mm £ 50 mm; 1.7mm
particle size). The reference standards of all the six
anthocyanins, i.e. petunidin chloride, pelargonidin
chloride, malvidin chloride, delphinidin chloride,
peonidin chloride and cyanidin chloride, were pur-
chasedfrom Extrasynthese SAS (Genay, France).
The fruit peel samples were extracted with 80:20
acetonitrile:water acidified with 0.3% phosphoric acid.
The samples were acid hydrolysed by the addition of
2M HCL at 1508C for 30 min in a sealed ampoule.
The flow rate was maintained at 0.5 ml min21, and the
temperature of the column and sample manager was
set at 408C and 58C, respectively. Injection volumes
were 2ml for standards as well as for samples. The
TUV detector was set at 525 nm and instrument
operations, data acquisition and processing were done
using EmPower2 chromatographic data software
(Waters Corporation). The anthocyanin peaks from
samples were identified by the comparison of retention
times with the corresponding retention times of
standards.
Results
Experiments designed by the Design-Expert software
as well as manual experiments were conducted to
optimize a solvent system providing high yield of
anthocyanins from the fruit peels of S. cumini.
Different solvents provided various colours to the
extracts (Figure 1) obtained from the ripe fruit skin
(Figure 1a).
Statistical analysis of response and levels
ANOVA of the quadratic regression model was used
for the statistical analysis. The adequacy of the model
was determined by evaluating the lack of fit, coefficient
Table IV. Anthocyanin content and colour values using solvents containing HCl.
Colour value
Sl. no. Ethanol (%) Acid (%) L* a* b* Chroma (C*) Hue angle (hab) Anthocyanin content mg 100 ml
21
1. 0 0.25 0.95 2.52 0.12 2.52 2.72 209.59
2. 0 0.50 0.87 2.23 0.06 2.23 1.54 226.68
3. 0 1.00 0.94 1.88 0.16 1.88 4.86 183.97
4. 20 0.25 1.40 3.40 20.63 3.45 349.51 153.20
5. 20 0.50 1.03 1.99 20.59 2.07 343.48 414.28
6. 20 1.00 1.16 3.22 20.11 3.22 358.04 218.17
7. 50 0.25 1.16 1.97 0.18 1.97 5.22 420.99
8. 50 0.50 1.06 2.50 20.12 2.50 357.25 277.50
9. 50 1.00 0.75 1.75 20.11 1.75 356.42 490.31
10. 80 0.25 0.76 1.77 0.30 1.79 9.61 622.06
11. 80 0.50 0.53 1.67 0.30 1.69 10.18 384.65
12. 80 1.00 0.63 2.34 0.14 2.34 3.42 234.59
13. 100 0.25 0.55 1.52 0.17 1.52 6.38 486.83
14. 100 0.50 0.50 1.61 0.38 1.65 13.28 464.97
15. 100 1.00 0.48 1.22 0.17 1.23 7.93 575.13
Table V. Anthocyanin content and colour values using solvents containing acetic acid.
Colour value
Sl. No. Ethanol (%) Acid (%) L* a* b* Chroma (C*) Hue angle (hab) Anthocyanin content mg 100 ml
21
1. 0 0.25 1.51 3.51 0.74 3.58 348.09 269.26
2. 0 0.50 0.77 2.23 0.12 2.23 3.08 347.90
3. 0 1.00 0.61 2.33 0.07 2.33 1.72 364.26
4. 20 0.25 1.74 4.07 1.48 4.33 340.01 430.24
5. 20 0.50 1.43 4.04 1.78 2.58 7.09 421.05
6. 20 1.00 0.84 2.57 0.32 4.41 336.22 763.80
7. 50 0.25 1.35 3.48 2.28 4.16 326.76 310.96
8. 50 0.50 1.39 3.69 2.34 4.36 327.61 634.60
9. 50 1.00 0.91 3.29 1.01 3.44 342.93 315.12
10. 80 0.25 1.31 2.57 22.00 3.25 322.10 274.76
11. 80 0.50 1.28 2.43 1.91 3.06 321.48 395.64
12. 80 1.00 1.40 2.89 22.40 3.75 320.29 281.53
13. 100 0.25 2.99 0.97 21.60 1.87 301.22 233.32
14. 100 0.50 2.63 1.70 2.27 2.83 306.82 238.88
15. 100 1.00 1.47 2.53 1.95 3.19 322.37 517.67
Solvent optimization for anthocyanin extraction from Syzygium cumini L. 367
of regression (R 2) and Fisher test value (F-value)
obtained from ANOVA. Statistical significance of the
model and model variables was determined at the 5%
probability level ( p , 0.05). The quadratic Equation
(3) was used by the software to build the response
surfaces. 3D response surface plots were generated by
keeping one response variable at its optimal level and
by plotting that against two factors (independent
variables). Table III shows the ANOVA summary for
the overall effect of factors on the three types of
response, i.e. the total anthocyanin content (Y1),
chroma value (Y2) and hue angle (Y3), as a function of
the three independent variables.
The R 2 values were 0.9630, 0.8104 and 0.9068,
and the coefficient of variance (CV) was 1.35%,
2.55% and 1.31% for Y1, Y2 and Y3, respectively. Very
low values of CV clearly indicated a very high degree of
precision and reliability of the experimental values.
The adequate precision that measures the signal-to-
noise ratio was .4 which shows an adequate signal-to-
noise ratio and indicates that this model can be used to
navigate the design space. The p value ,0.05 indicate
that the model terms are significant and the model
terms AB, AC, BC, A2, B2 and C2 were also
significant.
The diagnostics plots were also analysed to check
whether the model statistics were acceptable or not.
The residual analysis is necessary to confirm that the
assumptions for the ANOVA are met. The normal
probability plot of the studentized residuals was found
to be a straight line with the points falling on the line in
a linear fashion (Figure 2a) showing a normality of the
residuals. The predicted versus actual plot (Figure 2b)
was randomly scattered along the 458 line, and the
residuals versus predicted values (Figure 2c) were
randomly scattered which shows that the plots are
acceptable. This comparison of the experimental and
predicted values also confirmed the validity of the
model. Furthermore, the 3D mesh graph had the ideal
uniform round (Figure 2d,f) and square (Figure 2e)
shapes. The 3D mesh graph was plotted by taking one
response at a time and was compared with the acid
concentration and solvent concentration factors.
Table VI. Anthocyanin content and colour values using solvents containing citric acid.
Colour value
Sl. No. Ethanol (%) Acid (%) L* a* b* Chroma (C*) Hue angle (hab) Anthocyanin content mg 100 ml
21
1. 0 0.25 1.24 2.95 20.96 3.10 341.97 261.21
2. 0 0.50 1.05 2.97 20.56 3.02 349.32 238.62
3. 0 1.00 0.77 2.77 0.21 2.77 4.33 366.54
4. 20 0.25 1.30 3.15 21.02 3.31 342.05 379.28
5. 20 0.50 1.23 3.48 21.06 3.63 343.05 314.65
6. 20 1.00 1.10 3.66 20.78 3.74 347.96 255.92
7. 50 0.25 1.19 2.85 21.48 3.21 332.55 532.22
8. 50 0.50 1.12 2.83 21.32 3.12 334.99 494.60
9. 50 1.00 1.22 3.56 22.12 4.14 329.22 263.36
10. 80 0.25 2.00 1.84 22.33 2.96 308.29 263.69
11. 80 0.50 1.70 2.98 22.06 3.62 325.34 401.54
12. 80 1.00 1.39 3.23 22.47 4.06 322.59 327.52
13. 100 0.25 2.70 1.65 22.34 2.86 305.18 207.04
14. 100 0.50 2.43 0.97 20.93 1.34 316.21 239.82
15. 100 1.00 1.76 2.34 21.93 3.03 320.48 296.21
Figure 3. The 2D (a) and 3D (b) chromaticity plot of a*, b* and L* values of anthocyanin extracts.
B. Chaudhary and K. Mukhopadhyay368
Total anthocyanin content
The calculated total anthocyanin content of the
extracts using different solvents is shown in
Tables IV–VI. The highest anthocyanin content of
763.80 mg 100 ml21 was found when 20% ethanol was
used in combination with 1% acetic acid which is
much higher than other berries (Mazza and Miniati
1993). The control solvent water extract had a total
anthocyanin content of 290.38 mg 100 ml21. The
extracts obtained by solvents containing citric acid had
reduced amounts of total anthocyanins as compared
to HCl and acetic acid.
Colour characteristics
The anthocyanin extracts obtained using different
solvent systems provided varied range of chroma and
hue angle values. The hue angle values were in the
range of either 08 to 608 or 3008 to 3608 which have the
shades of red to yellow and pink to red, respectively
(Torskangerpoll and Anderson 2005). In terms of
highest chroma, again the solvent with 20% ethanol
and 1% acetic acid had the highest value of 4.1.
The colour measurement values and the determined
values of anthocyanin from different solvent systems
are shown in Tables IV–VI. The colour value for all
the 46 anthocyanin extracts is shown in 2D and 3D
chromaticity plots (Figure 3), which were obtained
using the software. It was clearly observed from the 2D
chromaticity plot (Figure 3a) that the extracts were
either in the pink range (positive a*) or in the blue
range (negative b*) and were having a low L* value
which implies a deep colour. The same kind of
observations were made in the 3D chromaticity plot
(Figure 3b) wherea* is red and 2a* is green in colour,
b* is yellow and 2b* is blue with L* and –L* values.
The control solvent had a chroma value of 1.48 and
hue angle 334.12.
Chromatographic analysis and antioxidant activity
The anthocyanin extracts of peels from the ripe fruits
of S. cumini were subjected to acid hydrolysis to obtain
the anthocyanidins to get an accurate quantification of
the target molecules in the samples by comparing with
the standards (Figure 4a) in this study. All the six
Figure 4. UPL chromatogram of (a) anthocyanin standard mixture and (b) S. cumini fruit peel extract.
Solvent optimization for anthocyanin extraction from Syzygium cumini L. 369
major types of anthocyanins such as cyanidin,
delphinidin, malvidin, petunidin, peonidin and pelar-
gonidin were identified in the sample (Figure 4b). The
total separation time was within 3 min.
The antioxidant activity of anthocyanins from fruit
peels extracted using the solvent prepared in 20%
ethanol in combination with 1% acetic acid was also
studied by quantifying the FRAP value. The fruit peel
extract showed an antioxidant activity of 4.34 ^ 0.26
Fe2þg21.
Discussion
This study aimed to optimize an extraction solvent to
get high yields of anthocyanin from S. cumini fruit
peels. Ethanol was used instead of methanol due to its
low toxicity and hence being safe for human
consumption. All the six major types of anthocyanins
were identified in the fruit peel samples by UPLC
studies. Pelargonidin has never been identified in this
fruit before (Brito et al. 2007; Veigas et al. 2007), and
the identification of all the six types of anthocyanin in
the fruits of S. cumini is being reported for the first
time to the best of our knowledge.
The manual and statistical solvent optimization
experiments suggest that the solvent percentage, acid
percentage and acid type significantly affect the
anthocyanin yield. The type of acid had little effect
on the yield statistically, but the manual experiments
suggest that acetic acid is the best acid to get high
anthocyanin from the fruit peel extracts of S. cumini.
The statistical analysis also indicates that the second-
order quadratic model employed in this study could be
used to optimize the anthocyanin extraction using
different sets of solvent. Earlier, it was reported that S.
cumini fruits have an anthocyanin content of 216 mg
100 ml21 where 0.1% HCl in methanol was used as
the extraction solvent (Veigas et al. 2007). The
optimized solvent has a low concentration of ethanol
and acetic acid which is safe for human consumption,
and the process can be used to extract high amounts of
anthocyanin (763.80 mg 100 ml21) from the S. cumini
fruit peels. Such high amount of anthocyanin yield has
not been reported from this fruit before. The relatively
high chroma (C* ¼ 4.41) and hue angle
(hab ¼ 336.22) were observed under the same con-
ditions. A positive correlation was found between the
amount of anthocyanin and chroma, the quantitative
component of chromaticity; a bidimensional par-
ameter was used to correlate the visual sensation that
attributes colourfulness (Yang et al. 2008).
The RSM was successfully employed to optimize
the extraction, and the three selected experimental
parameters have been evaluated in this study. The
model did not suggest any new combinations of
solvent for the anthocyanin extraction. All the three
variables were found to be significant for the extraction
of anthocyanins. The highest amount of anthocyanins
could be obtained if the numerical range for the
solvent concentration was set at 20–50% with a higher
acid percentage of acetic acid. These results had a
correlation with the manual results obtained and
hence the best solvent system for the anthocyanin
extraction, i.e. 20% ethanol with 1% of acetic acid
optimized by the manual experiments, was found to be
statistically correct.
The results obtained will be helpful for a better
utilization of S. cumini fruits in the food industries.
S. cumini fruits have been shown to have a high
antioxidant activity and was found suitable to be used
as a natural food colourant (Veigas et al. 2007).
Conclusions
A solvent system was optimized to get high yields of
anthocyanin from the peel extracts of S. cumini fruits.
The anthocyanin extracts of S. cumini fruits can also
be used as a food colourant like other fruits which are
sold commercially, but some more experimental
studies are required to confirm their suitability as
natural food colourants in the food processing
industries. The regression model developed by RSM
will help manufacturers optimize the extraction
process for high anthocyanin yield with minimum
use of the solvent.
Acknowledgements
The authors acknowledge Mr Ashwini Singh of
Central Instrumentation Facility of BIT, Mesra, for
operating the colorimeter.
Declaration of interest: The authors report no
conflicts of interest. The authors alone are responsible
for the content and writing of the paper. Bratati
Chaudhary is thankful to BIT, Mesra, Ranchi,
for providing fellowship and research facilities.
The authors are highly thankful to the Ministry of
Food Processing Industries (File No. 47/MFPI/
R&D/2006/517) and University Grants Commission
(F.No.34-275/2008-SR) for financial assistance.
References
Ahmed M, Akter MS, Eun JB. 2011. Optimization conditions for
anthocyanins and phenolics content extraction from purple sweet
potato using response surface methodology. Int J Food Sci Nutr
62:91–96.
Ayyanar M, Babu PS. 2012. Syzygium cumini (L.) Skeels: a review of
its phytochemical constituents and traditional uses. Asian Pac J
Trop Biomed 2:240–246.
Bas D, Boyaci IH. 2007. Modeling and optimization I: usability of
response surface methodology. J Food Eng 78:836–845.
Benzie IEF, Strain JJ. 1999. Ferric-reducing antioxidant power
assay: direct measure of total antioxidant activity of biological
fluids and modified version for simultaneous measurement of
total antioxidant power and ascorbic acid concentration.
Methods Enzymol 299:15–27.
Brito ESD, Araujo MCD, Alves RE, Carkeet C, Clevidence BA,
Novotny JA. 2007. Anthocyanins present in selected tropical
B. Chaudhary and K. Mukhopadhyay370
fruits: acerola, jambolao, jussara, and guajiru. J Agric Food
Chem 55:9389–9394.
Chalabi N, Gallon DJB, Vasson MP, Bignon YJ. 2008. Nutri-
genomics and antioxidants. Personalized Med 5:25–36.
Chaudhary B, Mukhopadhyay K. 2012a. Induction of anthocyanin
pigment in callus cultures of Solanum melongena L. in response to
plant growth regulators and light. IOSR J Pham 2:76–80.
Chaudhary B, Mukhopadhyay K. 2012b. Syzygium cumini (L.)
Skeels: a potential source of nutraceuticals. Int J Pharm Biol Sci
2:46–53.
CIE. 1986. Commission Internationale de I’E’clairage Colorimetry.
CIE Publication 15.2. 2nd ed., Vienna: Bureau Central of the CIE.
Delgado-Vargas F, Paredes-Lopez O. 2003. Natural colorants for
food and nutraceutical uses. Boca Raton, FL: CRC Press. p 326.
Grotewold E. 2006. The genetics and biochemistry of floral
pigments. Annual Rev Plant Biol 57:761–780.
Guisti MM, Wrolstad RE. 2003. Acylated anthocyanins from edible
sources and their applications in food systems. Biochem Eng J
14:217–225.
Huang CL, Liao WC, Chan CF, Lai YC. 2010. Optimization for the
anthocyanins extraction from purple sweet potato roots, using
response surface methodology. J Taiwan Agric Res 53:143–150.
Kapasakalidis PG, Rastall RA, Gordon MH. 2006. Extraction of
polyphenols from processed black currant (Ribes nigrum L.)
residues. J Agric Food Chem 54:4016–4021.
Lee J, Durst RW, Wrolstad RE. 2005. Determination of total
monomeric anthocyanins pigment content of fruit juices,
beverages, natural colourants and wines by the pH differential
method: collaborative study. J AOAC Int 88:1269–1278.
Mazza G, Miniati E. 1993. Anthocyanins in fruits, vegetables, and
grains. Boca Raton, FL: CRC Press.
Mortensen A. 2006. Carotenoids and other pigments asnatural
colorant. Pure Appl Chem 78:1477–1491.
Noonan JE. 1972. Color additives in food. In: Furia TE, editor.
Handbook of food additives. 2nd ed., West Palm Beach, FL:
CRC Press.
Torskangerpoll K, Anderson OM. 2005. Colour stability of
anthocyanins in aqueous solutions at various pH values. Food
Chem 89:427–440.
Veigas J, Narayan M, Laxman P, Neelwarne B. 2007. Chemical
nature, stability and bioefficacies of anthocyanins from fruit peel
of Syzygium cumini Skeels. Food Chem 105:619–627.
Vyas P, Chaudhary B, Mukhopadhyay K, Bandopadhyay R. 2009.
Anthocyanins: looking beyond colors. In: Bhowmik P, Basu SK,
Goyal A, editors. Advances in Plant Biotechnology. Oak Park, IL:
Bentham Science Publishers Ltd.
Yang Z, Fan G, Gu Z, Han Y, Chen Z. 2008. Optimization
extraction of anthocyanins from purple corn (Zea mays L.) cob
using tristimulus colorimetry. Eur Food Res Technol 227:
409–415.
Zou TB, Wang M, Gan RY, Ling WH. 2011. Optimization of
ultrasound-assisted extraction of anthocyanins from Mulberry,
using response surface methodology. Int J Mol Sci 12:
3006–3017.
Solvent optimization for anthocyanin extraction from Syzygium cumini L. 371
Copyright of International Journal of Food Sciences & Nutrition is the property of Taylor & Francis Ltd and its
content may not be copied or emailed to multiple sites or posted to a listserv without the copyright holder's
express written permission. However, users may print, download, or email articles for individual use.