Life Sciences | Agriculture Doi: 10.31276/VJSTE.63(2).

52-57

Extraction of flavonoids in pomelos’ peels

using Box-Behnken response surface design
and their biological activities
Truong Ngoc Quynh Phuong1, 2, Pham Van Hung1, 2, Nguyen Thi Lan Phi2, 3*
1
Department of Food Technology, International University, Vietnam National University, Ho Chi Minh city
2
Vietnam National University, Ho Chi Minh city
3
Department of Food Technology, University of Technology, Vietnam National University, Ho Chi Minh city
Received 9 January 2020; accepted 6 April 2020

Abstract:
The objectives of this study were to optimize the extraction conditions of flavonoids from the pomelo peel using the
Box-Behnken response surface methodology and to investigate the chemical composition and biological activities of
the extracts from three different pomelo species under optimized extraction conditions. Three extraction condition
factors of were varied, including the solid-solvent ratio (1/15-1/45 g/ml), temperature (30-60oC), and time (20-60
min), which affected the efficiency of the total flavonoid content (TFC). The high R2 coefficient of 0.93 indicated
that the experimental data obtained in this study fit well to a second-order polynomial using multiple regression
analysis. The maximum TFC extracted from the pomelo peel was 6.0 mg/g (dried basis, db), which was obtained
using Derringer’s desired function methodology under the following optimum conditions: a solid-solvent ratio of
1/44 g/ml, temperature of 30.7oC, and time of 34.6 min. The results also indicated that the Tan Trieu pomelo peel
had the highest TFC and naringin concentration followed by the Nam Roi and Duong Hong pomelo peels, whereas
the Nam Roi pomelo peel had a higher hesperidin concentration than the others. The extracts of the pomelo peels
exhibited strong antioxidant activities with low IC50 values. However, these extracts showed only a slight effect on
cancer cells, including lung cancer and breast cancer, when tested at 100 µg/ml.
Keywords: antioxidant, Box-Behnken design, extraction, flavonoids, pomelo.
Classification number: 3.1

Introduction and anti-inflammatory, anticancer, and antimicrobial

activities [1, 2]. In the pomelo peels, flavonoids are present
Pomelo (Citrus grandis), a member of the citrus family,
in three classes including flavanone, flavone, and flavanol
is grown in many eastern countries including India, Vietnam,
compounds, in which flavanone is considered as the
and Thailand. Pomelos have been used as fresh fruit or for
most abundant flavonoid in the pomelo peel (up to 80%).
juice processing with good quality and low price. However,
after processing, their peel is a primary by-product that Naringin has been reported to be a predominant flavanone
contributes to environmental pollution. Therefore, it is in the peel and edible portions of many varieties of pomelos
necessary to take full advantage of technology to develop [3-5], while a small concentration of hesperidin in a pomelo
other kinds of products from the peels [1]. In recent years, peel was also found [5, 6]. The naringin, hesperidin, and
pomelo has attracted more attention from scientists because neohesperidin contents were found to be much higher in
of their nutritional and antioxidant properties. In addition, the peels than in the juice [5]. Naringin has the ability to
pomelo peels and seeds were found to contain two natural prevent cancer by carcinogenesis suppression and induction
compounds such as flavonoids and limonoids, respectively, of cell apoptosis, whereas neohesperidin contributes to the
which play an important role in living systems. Significantly, bitter taste in grapefruit and pomelo, and has the power
flavonoids have a wide range of biological effects, such as to stimulate the immune system [2, 6-8]. Therefore, it is
chelation of metal catalysts, inhibition of key enzymes in necessary to extract these flavonoids from the peels of the
mitochondrial respiration, protection from heart disease, pomelo for food and pharmaceutical application.

*
Corresponding author: Email: [email protected]

Vietnam Journal of Science,

52 Technology and Engineering June 2021 • Volume 63 Number 2
 Life Sciences | Agriculture

Various methods have been applied to extract flavonoid Table 1. The coded level of variables chosen for the experiments.
compounds of plants based on the manipulation of the
physical properties of solvents to reduce surface tension, Range and level
Variable Coded
increase the solute’s solubility, promote the higher diffusion -1 0 +1
rate, and change in solvent polarity. Solid-liquid extraction
Solid-solvent ratio (w/v) X1 1/15 1/30 1/45
or the solvent extraction method, one of the most widely
used extraction methods, has been used to extract one or Incubation Temperature (oC) X3 30 45 60
more solutes from a solid matrix by a liquid solvent and has Incubation time (h) X4 30 45 60
widely applied in food industry to recover various products
including sugars, teas, coffees, vegetable oils, and functional Experimental design
compounds. The yield of the solutes is also influenced by
The total flavonoids of the pomelo peels were extracted
numerous factors including the preparation of the solid,
diffusion rate, temperature, and solvent choices [9]. Several with a solvent of 75% ethanol and optimized using the Box-
extraction methods have been reported for the extraction Behnken design with three variables including solid-solvent
of flavonoids from pomelo peels such as conventional ratio (1/15, 1/30 and 1/45), temperature (30, 45 and 60oC)
extraction, enzyme-assisted extraction, ultrasound-assisted and time (30, 45 and 60 min). The scientific basis behind
extraction, microwave-assisted extraction, supercritical the chosen testing levels was determined by screening
CO2 extraction, and pressurized fluid extraction [10-13]. experiments using the one-factor optimization method at
To the best of our knowledge, there is no report on the which the total flavonoids were the highest (the details of
optimization of extraction conditions to obtain maximum these experiments are not shown in this paper).
TFC from pomelo peels. Therefore, the objectives of this
study is to optimize extraction conditions such as solid- Each independent variable was coded at three levels of
solvent ratio, extraction temperature, and extraction time -1, 0, and +1 (Table 1). The experimental design consisted
to get the highest flavonoid from the pomelo’s peel based of a total of 17 experiments with five centre points as shown
on Box-Behnken response surface design and to determine in Table 2.
the chemical composition and biological properties of the
extracts from the Tan Trieu, Duong Hong, and Nam Roi Table 2. Total flavonoids content of pomelo peel extracted using
pomelo varieties. Box-Behnken design.

Materials and methods Trial no.

Solid-solvent ratio Temperature Time TFC
(w/v) (oC) (min) (mg/g)
Plant materials
1 1/15 45 20 4.31
Tan Trieu, Duong Hong, and Nam Roi pomelo (Citrus 2 1/30 45 40 5.81
grandis Osbeck) varieties grown at different locations of
Vietnam were used in this study. The pomelo fruits were 3 1/30 60 60 4.15
carefully washed to remove soil particles and dust. Peels 4 1/30 45 40 5.17
were taken out from the pulp and then cut into small pieces.
5 1/30 30 60 4.28
Then, the peel pieces were dried using a freeze-drying
machine and ground into fine powder. The pomelo peel 6 1/45 30 40 6.00
powder, with a moisture content of 12-14%, was stored in a 7 1/15 45 60 3.55
desiccator prior to the experiment.
8 1/45 60 40 5.26
Extraction method 9 1/45 45 20 4.94
The dried peel powder (1 g) was accurately weighed 10 1/30 30 20 5.21
and mixed with ethanol under predetermined extraction
11 1/15 60 40 4.81
conditions based on the Box-Behnken response surface
design as described in the experimental design section 12 1/30 45 40 5.81
below and in Table 1. After shaking the sample in a water 13 1/45 45 60 5.11
bath for a certain temperature and time, the samples were
14 1/30 60 20 4.67
centrifuged for 15 min at 4oC. The supernatant was kept
while the residue was repeatedly extracted 3 times. Then, 15 1/30 45 40 5.42
the supernatants were combined, evaporated, and diluted 16 1/15 30 40 4.94
with methanol to yield 20 ml of crude extract before storing
17 1/30 45 40 5.32
for analysis.

Vietnam Journal of Science,

June 2021 • Volume 63 Number 2
Technology and Engineering 53
Life Sciences | Agriculture

Determination of total flavonoids content the DPPH radical and the extracts at t=30 min.
The TFC was determined using the colorimetric method The results were reported as a half maximal inhibitory
previously described by Hung and Morita (2008) [14] with concentration (IC50) value. A lower IC50 value represents
a minor modification. The extract (0.5 ml) was mixed with stronger DPPH scavenging capacity. The IC50 value was
1.5 ml of 95% ethanol, 0.1 ml of 10% aluminium chloride determined from linear regression analysis using Microsoft
solution, and 0.1 ml of 1 M potassium acetate. Then, distilled Excel with data analysis add-in.
water was added to adjust the volume to 10 ml. The tubes
Cytotoxic activity on cancer cells
were thoroughly mixed and stood at ambient temperature
in the dark for 30 min. The absorbance was measured at The colorimetric cytotoxicity assay reported by Skehan,
415 nm against the reagent blank using a spectrophotometer et al. (1990) [16] was used to investigate the effects of
(Genesys 10S UV-Vis, USA). Rutin was used as the standard the extracts on the survival of cancer cells including
and the TFC was expressed as micrograms rutin equivalent Hep-G2 (human hepatocellular carcinoma), LU-1 (human
(RE) per gram of sample. lung adenocarcinoma), and MCF-7 (human breast
adenocarcinoma). By using SRB (sulforhodamine B) as a
HPLC analysis
basic colorimetric method, the cytotoxicity of the compounds
Flavonoids in the citrus peels were analysed using high was investigated. These cancer cells were maintained in
performance liquid chromatography (HPLC) according E’MEM (Dulbecco’s Modified Eagle Medium) with the
to the method of I.A. Ribeiro and M.H.L Ribeiro (2008) addition of L-glutamine, sodium pyruvate, NaHCO3, PSF
[15]. In detail, the extracts obtained at optimal extraction (penicillin-streptomycin sulfate-fungizone), NAA (non-
conditions were diluted with 20 ml of 0.02 M sodium acetate essential amino acids), and 10% BCS (bovine calf serum)
buffer (pH=4) and methanol (1:1). Then, the solutions were before storage at 37oC in a 5% CO2 incubator. A plate (96-
filtered through a 0.45 µm membrane filter before injection well) was used to grow cells at 104 cells/well for Hep-G2
into the HPLC system. The mobile phase, including and MCF-7, and 7.5×103 cells/well for LU-1 in the growth
acetonitrile (solvent A) and distilled water (solvent B), was medium. After 24 h of growth, these cells were incubated
used. The gradient elution was conducted as starting at 23% in the presence of the extracts with different concentrations
A in 8 min, 23-65% A in 7 min, 65-70% A in 5 min, 70% A in 48 h. Then, the total protein was maintained by using
- 23% A in 1 min, and completing the gradient at 23% A in trichloroacetic acid (Sigma) 50% and dyed by SRB 0.2%.
1 min. The elution was monitored at a flow rate of 1 ml/min The standard was measured by ellipticine, vinblastine, or
and UV wavelength of 280 nm. Naringin (Product #N1376, taxol dissolving DMSO. The result was read by an ELISA at
Sigma), hesperidin (product #H5254), and HPLC grade 495-515 nm. The percentage of cell survival is determined
acetonitrile (product #1000302500) and methanol (product by:
#1060182500, Merck) were used in this study.
% cell survival=(Abssample-Abscontrol)/(AbsDMSO-Abscontrol)
DPPH radical scavenging assay
Statistical analysis
The antioxidant activity of the extracts was determined
The data were analysed for multiple regression analysis
according to the method of Hung and Morita (2008) [14].
and analysis of variance (ANOVA) to fit the mathematical
The final concentration of the DPPH solution was 0.075
modelling using Design Expert software (Version 11,
mM. A mixture of 3.9 ml of DPPH solution and 0.1 ml
Stat-Ease Inc., USA). After fitting the data to the models,
of the extract was mixed and kept in the dark at ambient
the response surface and 3D contour were plotted and
temperature for 30 min. Then, the absorbance of the mixture
investigated.
was read at 515 nm. Methanol (0.1 ml) was used to replace
the extract to mix with the DPPH solution and counted as Results and discussion
the blank sample. The scavenging of DPPH was calculated
Box-Behnken design analysis
according to the following equation:
The statistically designed experiments under different
%cDPPHcscavenging={[Abs(t=0)-Abs(t=30)]/Abs(t=0)}×100
extraction conditions were carried out to investigate the
where Abs(t0) is the absorbance of the DPPH radical and combined effect of independent variables (solid-solvent
methanol solution at t=0 min and Abs(t30) is absorbance of ratio, extraction temperature and extraction time) on the

Vietnam Journal of Science,

54 Technology and Engineering June 2021 • Volume 63 Number 2
 Life Sciences | Agriculture

extraction yield of the TFC and the results are shown in Fitting of second order polynomial equation
Table 2. The results indicated that the TFC of the extracts The relationship between the three factors with the
from the pomelo peel varied with different extraction response is given by the following equation in terms of
conditions. Table 3 showed the fitting of four high degree coded factors:
polynomial models (linear, interactive (2FI), quadratic, and TFCv(mg/g)=5.51+0.4625X 1 -0.1925X 2 -0.2550X 3 -
cubic models) with the experimental data. The quadratic 0.1525X 1 X 2 +0.2325X 1 X 3 +0.1025X 2 X 3 -0.1768X 1 2 -
model had a maximum adjusted R2 with low p-value. 0.0768X22-0.8517X32
Therefore, this model was the most suitable for the designed
where X1 is the solid-solvent ratio, X2 is the extraction
experiments. The experimental data was analysed by
temperature, and X3 is the extraction time.
ANOVA and the results in Table 4 indicated that the model
was significant because the p-value of the model was less The results indicated that the TFC of the extracts from
than 0.05. Moreover, the F-value was 0.6093 meaning that the pomelo peels were not only dependent on all factors (X1,
X2, and X3), but also dependent on the relationship between
the lack of fit of the model was not significant as compared
the two independent factors (X1X2, X1X3, and X2X3).
to pure error and the designed model was good.
Effect of independent variables on the TFC
Table 3. Sequential model fitting for the response.
Three factors including the solid-solvent ratio, extraction
Sequential Lack of fit temperature, and time at three different levels were used to
Source Adjusted R2
p-value p-value investigate the influence of the independent variables in
Linear 0.095199 0.067263 0.232767 the TFC. The 3D response surface and contour plots were
developed from the model by maintaining two factors as
2FI 0.820918 0.041751 0.086334
constant while varying the third to illustrate the optimum
Quadratic 0.001709 0.64349 0.831442 Suggested conditions (Fig. 1).
Cubic 0.64349 0.797545 Aliased Effect of solid-solvent ratio: Figs. 1B and 1C showed
that the TFC increased with increasing solid-solvent ratio
Table 4. Analysis of variance (ANOVA) for Box-Behnken model. from 1:15 to 1:44 (g/ml). The difference in concentration
between the cell tissue of the pomelo peel and the solvent
Source
Sum of Degree of Mean
F-value P-value
improved the efficiency of extraction because of an increased
squares freedom square mass transfer rate. The liquid circulation and turbulence
Model 6.22 9 0.6908 9.77 0.0033 produced by cavitation increased the contact surface
between the solvent and target compounds by permitting
Lack-of-fit 0.1552 3 0.0517 0.6093 0.6435 greater penetration of solvent into the sample matrix [17].
Pure error 0.3397 4 0.0849 Effect of extraction temperature: the effects of extraction
Corrected total 6.71 16 temperature on the TFC under the Box-Behnken design can
be observed in Figs. 1A and 1C. The results indicated that
R2 0.9263
temperature did not affect TFC at solid-solvent ratio of
adj R 2
0.8314 1:15 g/ml while it did affect the TFC at higher solid-solvent
ratios. As a result, the TFC reached its highest level when
C.V% 6.71%
the temperature was 30oC and the solid-solvent ratio was

Fig. 1. Response surface plots representing the effect of extraction conditions on the TFC. a: the solid-solvent ratio; b: extraction
temperature; c: time at three different levels.

Vietnam Journal of Science,

June 2021 • Volume 63 Number 2
Technology and Engineering 55
Life Sciences | Agriculture

1:44 g/ml. As a result, the 30oC temperature was the ideal In the present study, the concentration of two flavonoids
temperature for extraction of TFC from the pomelo peels in in pomelo peels including naringin and hesperidin was
this study. identified using the HPLC and are given in Table 5. The
Effect of extraction time: the effects of extraction time on naringin concentration in the Tan Trieu pomelo was the
the TFC under the Box-Behnken design are shown in Figs. highest (2.2±0.2 mg/g peels) followed by the Duong
1A and 1B. The TFC increased with increasing extraction Hong pomelo peel (1.6±0.1 mg/g peel) and the Nam Roi
time from 20 to 45 min. At a higher solid-solvent ratio, the pomelo peel (0.2±0.2 mg/g peel). The highest hesperidin
TFC reduced upon lengthening the extraction time. The concentration was 0.69±0.3 mg/g found in the Nam Roi
highest TFC of the pomelo peels was obtained with an pomelo peel, followed by the Tan Trieu and Duong Hong
extraction time of 34.6 min. Further increasing extraction pomelo peels. The results found in this study were consistent
time could cause the loss of solvent by vaporization. with the results reported by Xu, et al. (2008) [20], which
stated that the hesperidin concentration in the pomelo peels
Optimization and verification of model
was 1.77 mg/g DW under the same extraction conditions.
The extraction conditions were optimized to get the However, hesperidin could not be found in several of the
maximum extractive TFC by employing Derringer’s desired pomelo types from China. It has been shown that flavonoids
function methodology. From the model, the highest TFC are not equally distributed among the different types and
was predicted to be 6.0 mg/g under the following optimized parts of the citrus fruit. Jang, et al. (2010) [21] reported
conditions: the solid-solvent ratio of 1/44 g/ml, extraction that the essential oil of the Buntan peel contained a higher
temperature of 30.7oC, and extraction time of 34.6 min. TFC than those from the extracts of fruit pulp with different
For confirmation, triplicate extractions of TFC from solvents. The greatest TFC among the different parts of the
pomelo peels were carried out under the optimum conditions fruits ranked as follows: peel>pulp>juice [3].
stated above. The average TFC was 6.0±0.1 mg/g peels, DPPH radical scavenging activities of pomelo peel
which is not significantly different from the predicted extracts
results of the model. Thus, the model obtained in this study
was significant. The TFC of the pomelo peels extracted The antioxidant capacities of the pomelo peel extracts
under the optimal conditions in this study was significantly measured by the DPPH scavenging activity and expressed
higher than that obtained by the conventional enzyme and as IC50 values are given in Table 6. The IC50 value was
ultrasound-assisted extraction methods reported by Hung, calculated as the amount of extract required to inhibit 50%
et al. (2020) [18]. of the DPPH radical. The lower the IC50 value is, the higher
the antioxidant activity of the extract. The IC50 values of the
Flavonoid concentration in pomelo’s peels
citrus peel extracts ranged from 1.08 to 4.51 μg/ml, which
The TFCs of the peel extracts from different pomelo is significantly higher than that of the Trolox (0.65±0.04
varieties are given in Table 5. The TFC varied among μg/ml). These results indicate that the Tan Trieu pomelo
the different varieties of pomelo (p<0.05), which ranged peel extract exhibited the highest antioxidant activity
from 4.3±0.1 to 6.0±0.1 mg/g peels. The lowest TFC was (IC50=1.08±0.04 μg/ml), followed by the Nam Roi pomelo
presented in the Duong Hong pomelo peels (4.3±0.1 mg/g peel extract (IC50=3.09±0.01 μg/ml) and Duong Hong
peels), whereas the highest TFC was 6.0±0.1 mg/g peels pomelo peel extract (IC50=4.51±0.03 μg/ml). Previous
from the Tan Trieu pomelo peels. The difference in the TFC studies [22] revealed that the IC50 of ethanolic extracts
among the different pomelo varieties might be due their from the peels of C. hystrix from Boyolali - Central Java,
chemical composition, maturity at harvest, soil and water Indonesia were 16.7 μg/ml. The present study [23] also
qualities, and the condition of the post-harvest method [19]. indicated that the peels of the pomelo varieties in Vietnam
Table 5. Total flavonoid, naringin and hesperidin concentrations had a higher antioxidant capacity than other locations.
of pomelo peels’ extracts under optimized condition (mg/g).
Table 6. DPPH radical scavenging (IC50 value) of pomelo peels’
Concentration (mg/g, db) extracts.
Sample
Total flavonoids Naringin Hesperidin Sample IC50 value (µg/ml)
Trolox 0.65±0.04a
Tan Trieu pomelo 6.0±0.1b 2.2±0.2c 0.36±0.1b
Tan Trieu pomelo 1.08±0.04b
Duong Hong pomelo 4.3±0.1a 1.6±0.1b 0.18±0.2a
Duong Hong pomelo 4.51±0.03d
Nam Roi pomelo 4.5±0.1 a
0.2±0.2a
0.69±0.3c
Nam Roi pomelo 3.09±0.01c
: different letters in the same column are significantly
a, b, c
: different letters in the same column are significantly
a, b, c

different (p<0.05). different (p<0.05).

Vietnam Journal of Science,

56 Technology and Engineering June 2021 • Volume 63 Number 2
 Life Sciences | Agriculture

Cytotoxic activity on cancer cells of peel’s extracts REFERENCES

Table 7. Anticancer activity of pomelo peels’ extracts. [1] J.B. Harborne, C. William (2000), “Advances in flavonoids research since
1992”, Photochem., 55(6), pp.481-504.
Initial concentration CS value (%) [2] M.K.N. Chidambara, et al. (2012), “Differential inhibition of human colon
Sample cancer cells by structurally similar flavonoids of citrus”, Food Chem., 132(1), pp.27-34.
(µg/ml) Hep-G2 LU-1 MCF-7
Standard (+) 5 4.34±0.42a 1.13±0.44a 2.53±0.06a [3] Y. Nogata, et al. (2006), “Flavonoid composition of fruit tissues of citrus
species”, Biosci. Biotechnol. Biochem., 70(1), pp.178-192.
Tan Trieu 100 98.94±0.95 b
99.24±1.06 c
80.44±1.92 c
[4] M. Zhang, et al. (2011), “The flavonoid composition of flavedo and juice from
Duong Hong 100 97.42±1.13b 83.77±1.51b 99.54±0.62d the pummelo cultivar (Citrus grandis L. Osbeck) and the grapefruit cultivar (Citrus
Nam Roi 100 99.04±0.79b 85.42±0.52b 77.11±2.06b paradisi) from China”, Food Chem., 129(4), pp.1530-1536.
a, b, c
: different letters in the same column are significantly different [5] Y.C. Wang, et al. (2007), “Quantitation of bioactive compounds in citrus fruits
(p<0.05). cultivated in Taiwan”, Food Chem., 102(4), pp.1163-1171.
[6] M.A. Alam, et al. (2014), “Effect of citrus flavonoids, naringin and naringenin,
The anticancer activities of the pomelo peel extracts are on metabolic syndrome and their mechanisms of action”, Adv. Nutr., 5(4), pp.404-417.
shown in Table 7. The results indicate that the pomelo peels [7] O. Benavente-Garcia, J. Castillo (2008), “Update on uses and properties of
did not affect Hep-G2 (human hepatocellular carcinoma). citrus flavonoids: new findings in anticancer, cardiovascular, and anti-inflammatory
The percentage of dead cells after treatment was 1.06% (Tan activity”, J. Agric. Food Chem., 56(15), pp.6185-6205.
Trieu pomelo peel extract), 2.58% (Duong Hong pomelo peel [8] I. Erlund (2004), “Review of the flavonoids, quercetin, hesperidin, and
extract), and 0.96% (Nam Roi pomelo peel extract). The Tan naringenin. Dietary sources, bioactivities, bioavailability, and epidemiology”, Nutr.
Res., 24(10), pp.851-874.
Trieu pomelo peel extract also had no effect on Lung cancer
cells (LU-1), in which 99.24% of the cancer cells survived. [9] D.V. Dandekar, et al. (2008), “Hydrotropic extraction of bioactive limonin
from sour orange (Citrus aurantium L.) seeds”, Food Chem., 109(3), pp.515-520.
The Duong Hong and Nam Roi pomelo peel extracts slightly
affected the LU-1 cells with a total percentage of dead cancer [10] S.M. Jeong, et al. (2004), “Effect of heat treatment on the antioxidant activity
of extracts”, J. Agric. Food Chem., 52(11), pp.3389-3393.
cells being 16.23% and 14.58%, respectively. In addition, the
highest percentage dead of breast cancer cells (MCF-7) was [11] Y. Ma, et al. (2009), “Simultaneous extraction of phenolic compounds of
citrus peel extracts: effect of ultrasound”, Ultrason. Sonochem., 16(1), pp.57-62.
22.89% when treated with the Nam Roi pomelo peel extract,
followed by the Tan Trieu pomelo peel extract (19.56%) and [12] E.M. Garcia-Castello, et al. (2015), “Optimization of conventional and
ultrasound assisted extraction of flavonoids from grapefruit (Citrus paradisi L.) solid
Duong Hong pomelo peel extract (0.46%). The previous study wastes”, LWT-Food Sci. Technol., 64(2), pp.1114-1122.
[24] also reported that the essential oils of lemon and grapefruit [13] N. M’hiri, et al. (2014), “Extraction methods of citrus peel phenolic
peels exhibited a weak cytotoxicity toward human prostate compounds”, Food Rev. Int., 30(4), pp.265-290.
(PC-3), lung (A549), and breast (MCF-7) tumour cell lines. [14] P.V. Hung, N. Morita (2008), “Distribution of phenolic compounds in the
graded flours milled from whole buckwheat grains and their antioxidant capacity”,
Conclusions Food Chem., 109(2), pp.325-331.
The TFC of pomelo peel extracts were successfully [15] I.A. Ribeiro, M.H.L. Ribeiro (2008), “Naringin and naringenin determination
maximized using the Box-Behnken design and response and control in grapefruit juice by a validated HPLC method”, Food Control, 19(4),
pp.432-438.
surface analysis. The results indicated that the solid-solvent
ratio, extraction temperature, and extraction time were the [16] P. Skehan, et al. (1990), “New colorimetric cytotoxicity assay for anticancer-
drug screening”, J. Natl. Cancer Inst., 82(13), pp.1107-1112.
most important factors that affect TFC values. After analysis,
[17] M. Romdhane, C. Gourdon (2002), “Investigation in solid-liquid extraction:
the quadratic models within the studied experimental range Influence of ultrasound”, Chem. Eng. J., 87(1), pp.11-19.
of the various process variables were used to maximize the
[18] P.V. Hung, et al. (2020), “Chemical composition and biological activities
TFC. As a result, the optimized TFC of the pomelo peel was of extracts from pomelo peel by-products under enzyme and ultrasound-assisted
6.0 mg/g peels under the optimized extraction conditions, extractions”, J. Chem., Doi: 10.1155/2020/1043251.
just as predicted. All pomelo peel extracts possessed strong [19] C. Shahzad (2011), “Total phenolic content and antioxidant capacity of
antioxidant activities with a low IC50. However, the pomelo extracts obtained from 6 important fruit residues”, Food Res. Int., 44(1), pp.391-396.
peel extracts showed weak or moderate anticancer activities [20] G.H. Xu, et al. (2008), “Minerals, phenolic compounds, and antioxidant
when treated with 100 μg/ml of the extract. capacity of citrus peel extract by hot water”, J. Food Chem., 73(1), pp.11-17.
[21] H.D. Jang, et al. (2010), “Antioxidant potentials of buntan pumelo (Citrus
ACKNOWLEDGEMENTS grandis Osbeck.) and its ethanolic and acetified fermentation products”,  Food
Chem., 118(3), pp.554-558.
This research is funded by Vietnam National Foundation
for Science and Technology Development (NAFOSTED) [22] I. Fidrianny, et al. (2015), “Antioxidant activities in various peel extracts of
four varieties rambutan (Nephelium lappaceum) using DPPH, FRAP assays”, Int. J.
under grant number 106-NN.02-2016.72. Pharmacog. Phytochem. Res., 7(2), pp.280-285.
COMPETING INTERESTS [23] W. Xi, et al. (2014), “Flavonoid composition and antioxidant activities of
Chinese local pummelo (Citrus grandis Osbeck.) varieties”, Food Chem., 161, pp.230-238.
The authors declare that there is no conflict of interest [24] Y. Zu, et al. (2010), “Activities of ten essential oils towards propionibacterium
regarding the publication of this article. acnes and PC-3, A-549 and MCF-7 cancer cells”, Molecules, 15(5), pp.3200-3210.

Vietnam Journal of Science,

June 2021 • Volume 63 Number 2
Technology and Engineering 57