foods

Article
Application of Aqueous Saline Process to Extract Silkworm
Pupae Oil (Bombyx mori): Process Optimization and
Composition Analysis
Janjira Tangsanthatkun 1,2 , Methavee Peanparkdee 1,2 , Wattinee Katekhong 1,2 , Thepkunya Harnsilawat 2,3 ,
Chin Ping Tan 4 and Utai Klinkesorn 1,2, *

1 Department of Food Science and Technology, Faculty of Agro-Industry, Kasetsart University, 50 Ngam Wong
Wan Road, Chatuchak, Bangkok 10900, Thailand; [email protected] (J.T.); [email protected] (M.P.);
[email protected] (W.K.)
2 Research Unit on Innovative Technologies for Production and Delivery of Functional Biomolecules, Kasetsart
University Research and Development Institute (KURDI), 50 Ngam Wong Wan Road, Chatuchak,
Bangkok 10900, Thailand; [email protected]
3 Department of Product Development, Faculty of Agro-Industry, Kasetsart University, 50 Ngam Wong Wan
Road, Chatuchak, Bangkok 10900, Thailand
4 Department of Food Technology, Faculty of Food Science and Technology, Universiti Putra Malaysia,
Serdang 43400, Selangor, Malaysia; [email protected]
* Correspondence: [email protected]

Abstract: Silkworm pupae, a waste product from the silk production industry, can be an alternative
source of edible oil, thus reducing the industry’s waste. In the present work, frozen silkworm pupae
were used as raw material to extract oil via an aqueous saline process. The Box–Behnken design



(BBD) and response surface methodology (RSM) were used to optimize the extraction process. The

 extraction conditions with the highest oil yield and a low peroxide value were obtained when using a
Citation: Tangsanthatkun, J.; saline solution concentration of 1.7% w/v, a ratio of aqueous liquid to silkworm pupae of 3.3 mL/g,
Peanparkdee, M.; Katekhong, W.; and a 119 min stirring time at the stirring speed of 100 rpm. Under these conditions, silkworm oil with
Harnsilawat, T.; Tan, C.P.; Klinkesorn, a yield of 3.32%, peroxide values of approximately 1.55 mM, and an acid value of 0.67 mg KOH/g oil
U. Application of Aqueous Saline
was obtained. The extracted oil contained omega-3 acids (α-linolenic acid), which constituted around
Process to Extract Silkworm Pupae
25% of the total fatty acids, with approximate cholesterol levels of 109 mg/100 g oil. The amounts of
Oil (Bombyx mori): Process
β-carotene and α-tocopherol were approximately 785 and 9434 µg/100 g oil, respectively. Overall,
Optimization and Composition
the results demonstrated that oil extracted from silkworm pupae has good quality parameters and
Analysis. Foods 2022, 11, 291.
https://doi.org/10.3390/
thus can be used as a new valuable source of edible lipids.
foods11030291
Keywords: silkworm pupae; oil extraction; aqueous saline method; sodium chloride concentration;
Received: 13 December 2021
response surface methodology; fatty acid profile; chemical composition
Accepted: 19 January 2022
Published: 21 January 2022

Publisher’s Note: MDPI stays neutral

with regard to jurisdictional claims in 1. Introduction
published maps and institutional affil-
Silkworm pupae are the by-product of the commercial silk industry. China, India,
iations.
Uzbekistan, Thailand, and Brazil are the top five silk-producing countries in the world [1].
The desilked silkworm pupae account for 60% of the dry cocoon weight, and they are
mostly used as fertilizer and animal feed, or are regarded as industrial waste [2]. Approx-
Copyright: © 2022 by the authors.
imately 400,000 metric tons of dry silkworm cocoon are obtained from the production
Licensee MDPI, Basel, Switzerland.
of 125,000 metric tons of raw silk [3]. During 2015–2020, the average annual global silk
This article is an open access article production was approximately 155,400 metric tons based on statistics from the International
distributed under the terms and Sericultural Commission (https://inserco.org/en/statistics: accessed on 10 January 2022).
conditions of the Creative Commons Thus, approximately 497,300 metric tons of dry cocoon were produced, which equates
Attribution (CC BY) license (https:// to approximately 298,400 metric tons of desilked pupae each year. Recently, the use of
creativecommons.org/licenses/by/ agricultural waste and the management of industry by-products have become a focus of
4.0/). research as we seek to curb climate change and minimize our impact on the Earth [4]. With

Foods 2022, 11, 291. https://doi.org/10.3390/foods11030291 https://www.mdpi.com/journal/foods

Foods 2022, 11, 291 2 of 16

this in mind, the proper utilization of silkworm pupae could not only generate extra income
for farmers in the silk industry but could significantly reduce the industry’s inherent waste.
In Thailand, silkworms are classified as mulberry (Bombyx mori) and non-mulberry or eri
silkworm (Samia ricini), each of which has different nutritional value. Two main nutrients
in the pupae of silkworms are protein and oil, which vary in the range of 48–67 wt% and
17–30 wt% (dry basis), respectively [5].
Oil extracted from silkworm pupae has been reported as a safe oil source and nutri-
tionally equivalent to commonly used vegetable oils, such as sunflower oil [6]. This oil is a
source of unsaturated fatty acids (approximately 60–70% of the total fatty acid content),
particularly α-linolenic and oleic acids. Both α-linolenic and oleic acids are known for
their nutritional and health benefits, which could be exploited for various applications,
including food, supplements, and feed [2,7–9]. To separate oil from the silkworm pupae,
several techniques were previously used, including mechanical pressing extraction [10]
and solvent and supercritical fluid extraction [2,6,11–14]. Using these methods requires
the silkworm pupae to be dry before the extraction can be initiated, and the extraction
conditions may alter the quality of the protein co-products. Due to its safe nature and
limited environmental impact, aqueous oil extraction can be used as an alternative method
for the extraction of oil from silkworm pupae.
In the aqueous extraction method of lipids from insects or the muscle tissue of other or-
ganisms (e.g., fish), mincing, grinding, or homogenizing is usually done in the procedure’s
early steps. Next, the lipids in grounded or homogenized tissues are then extracted with an
aqueous solution using tissue hydrolysis with enzymes, alkaline, or acids to improve the
extraction efficiency [15–19]. However, using a hydrolysis reaction limits the potential reuse
of protein components and regularly results in a bitter taste [20,21]. Moreover, issues have
been reported with attempts to scale up the enzymatic hydrolysis method [22]. In previous
studies, common salt or sodium chloride has been shown to enhance the oil extraction from
olive paste and fish tissue by increasing the gradient of ionic charge and density between
the oil and hydrophilic phases [22,23]. Using an aqueous saline extraction process to extract
the oil from silkworm pupae has not yet been investigated and could prove to be a valuable
extraction method.
This study aimed to develop an aqueous saline method for extracting oil from silk-
worm pupae. The main variables investigated in the use of this method included deter-
mining the ideal sodium chloride concentration, aqueous liquid-to-solid (silkworm) ratio,
and stirring time. Response surface methodology (RSM) is an optimal tool for process
optimization. Hence, we used RSM to optimize the extraction conditions in order to en-
hance the production yield and the quality of the silkworm oil. In addition, the fatty acid
composition and other components, including cholesterol, α-tocopherol, and β-carotene, of
the extracted silkworm oil were also analyzed.

2. Materials and Methods

2.1. Raw Materials and Chemicals
Frozen mulberry desilked silkworm pupae (Bombyx mori), used in all the experi-
ments, were obtained from a local silkworm farm in Phetchabun Province, Thailand. The
silkworm pupae samples were kept at approximately −18 ◦ C until the experiment was
performed. Deionized water (electricity resistivity ≥ 15 MΩ.cm) was used for the prepa-
ration of all the solutions. Sodium chloride (NaCl), phenolphthalein (C20 H14 O4 ), barium
chloride (BaCl2 ), iron (II) sulfate (FeSO4 ), and ammonium thiocyanate (NH4 SCN) were pur-
chased from Ajax Finechem Pty, Ltd. (New South Wales, Australia). Hexane (C6 H14 ) and
butanol (C4 H10 O) were purchased from RCI Labscan, Ltd. (Bangkok, Thailand), while iso-
propyl alcohol (2-propanol) was purchased from J.T. Baker (Avantor Performance Materials,
LLC, Allentown, PA, USA). Toluene (C7 H8 ), potassium hydroxide (KOH), and methanol
(CH3 OH) were supplied by Merck KGaA (Darmstadt, Germany). Cumene hydroperoxide
(C6 H5 C(CH3 )2 OOH) and 2,2-diphenyl-1-picrylhydrazyl (DPPH) were purchased from
Sigma-Aldrich Co. (St. Louis, MO, USA). All the chemicals used were reagent-grade.
Foods 2022, 11, 291 3 of 16

2.2. Extraction of the Silkworm Pupae Oil

The aqueous saline extraction process was carried out according to the method of
Tzompa-Sosa et al. [19] and Kadioglu et al. [24], with some modifications. To extract the oil,
100 g of each frozen silkworm pupae was mixed with a saline (NaCl) solution of a specified
concentration and volume. Then, the mixture was blended for 3 min using a food processor
(Phillips HR2118, Phillips International Co., Ltd., Samut Prakan, Thailand), followed by
stirring at 100 rpm with an overhead stirrer (IKA® Rw 20 digital, Staufen, Germany) for a
specified time. Next, the suspension was passed through a 45-mesh stainless-steel screen
to eliminate large particles (>354 µm). The obtained silkworm pupae suspension was
then centrifuged (Sorvall RC6 plus, Thermo Fisher Scientific, Karlsruhe, Germany) at
10,000 rpm (~15,180× g) for 1 h at 25 ◦ C to obtain three fractions, including a cream layer,
a supernatant or an aqueous layer, and a pellet from top to bottom, respectively. The top
cream layer and supernatant phase were collected and subsequently centrifuged using a
high-speed, refrigerated microcentrifuge (MX-307, TOMY Digital Biology Co., Ltd., Tokyo,
Japan) at 15,000 rpm (~20,380× g) for 20 min at 25 ◦ C. This second centrifugation was
performed at a higher speed than the first one to enhance lipid separation. After the second
centrifugation, free oil was obtained in the upper layer of the centrifuge tube. The lower
layers contained a thin emulsion layer, a clear supernatant, and a residue. The transparent
oil and emulsion layers were then pipetted and mixed with hexane (30 mL). The extracted
oil was separated by an evaporating solvent using a parallel evaporator equipped with a
vacuum controller (Buchi Syncore® , BÜCHI Labortechnik AG Co., Flawil, Switzerland) at
50 ◦ C and 250 mbar. Nitrogen flushing was also used to ensure that there was no solvent
remaining in the extracted oil and to replace the oxygen, which can affect the quality of the
oil. The amount of extracted oil was weighed to calculate the extraction yield by following
Equation (1). The extracted oil for each treatment was kept in an amber glass bottle and
stored at approximately −18 ◦ C for further analysis.

Extraction yield (%) = (Oil weight/Sample weight) × 100 (1)

2.3. Determination of the Peroxide Value (PV)

The PV of the silkworm pupae oil was analyzed by following the lipid peroxide assay
adapted from Prichapan et al. [25]. Briefly, 1 mL of a barium chloride solution (0.132 M
BaCl2 in 0.4 N HCl) was added to 1 mL of an iron sulfate solution (0.144 M FeSO4 ), then vor-
texed for 1 min, and centrifuged (Gallenkamp Junior Centrifuge, Gallenkamp & Co., Ltd.,
London, UK) at a maximum speed of 4460 rpm (~2450× g) for 3 min. Next, 1 mL of the
clear supernatant was taken and mixed with 1 mL of an ammonium thiocyanate solution
(3.94 M NH4 SCN) to obtain the ferrous reagent. Next, 30 µL of this reagent solution was
added to the oil sample solution, which was previously prepared by dissolving 0.2 mL oil
in a 2.8 mL methanol/butanol mixed solution (2:1 v/v). The sample solution was kept in
the dark at room temperature for 20 min, and the absorbance of the resulting solution was
then measured at a wavelength of 510 nm using a UV–Vis spectrophotometer (Genesys
20, Thermo Spectronic, Waltham, MA, USA). The lipid hydroperoxides or PV in mM were
calculated using the standard curve of cumene hydroperoxide.

2.4. Determination of the Acid Value (AV)

The AV of the silkworm pupae oil was determined by following the AOCS Cd-3d-63
method [26], where 1 g of oil was dissolved in an isopropyl alcohol and toluene solution
(1:1 v/v). After mixing, the sample solution was titrated using a 0.01 N potassium hydrox-
ide (KOH) solution and phenolphthalein as an indicator. The exact volume of the titrant
was recorded when the end point (pink solution) was reached. The AV in mg KOH/g oil
was expressed after calculations according to the following equation:

AV = [(A − B) × N × 56.1]/W (2)

Foods 2022, 11, 291 4 of 16

where A and B represent the mL of KOH solution used in the titrating sample and the
blank, respectively. N is the normality of the KOH solution, and W is the mass of the oil
sample (g).

2.5. Analysis of the Chemical Composition of the Silkworm Pupae and Oil
The chemical composition of the silkworm pupae, including the moisture, total protein,
total fat, total carbohydrate, dietary fiber, and ash, was determined according to the method
of the Association of Official Analytical Chemists (AOAC). The fatty acid composition of
the oil was analyzed using gas–liquid chromatography (GLC) according to the in-house
method based on the official AOAC 963.22 and AOAC 969.33 methods. The standard AOAC
994.10 analytical method was used to analyze the cholesterol content [27]. To determine
the α-tocopherol and β-carotene content, in-house methods using high-performance liquid
chromatography (HPLC) were applied [28,29].

2.6. Determination of 2,2-Diphenyl-1-Picrylhydrazyl (DPPH) Free Radical Scavenging Activity

The DPPH radical scavenging activity of silkworm oil was determined using the
method of Singh et al. [30], with slight modifications. An aliquot (100 µL) of oil was
added to 4.9 mL of a 0.1 mM DPPH solution in methanol. The solution was then vortexed
vigorously and left for 30 min in the dark under ambient conditions. The absorbance was
measured at 517 nm using a UV–Vis spectrophotometer (Genesys 20, Thermo Spectronic,
Waltham, MA, USA). The absorbance of each sample was recorded, and the percentage
inhibition of DPPH was calculated according to the following formula:

Inhibition percentage (%) = 100 × (AB − AS )/AB (3)

where AB and AS are the absorbance values of the blank (methanol) and of the tested
samples, respectively.

2.7. Experimental Design and Statistical Analysis

Three independent variables, namely the concentration of NaCl (X1 , % w/v), the
liquid-to-solid ratio (X2 , mL/g), and the stirring time (X3 , min), were investigated for
the aqueous saline extraction of silkworm pupae oil. In the Box–Behnken design (BBD),
three levels for each variable were applied according to the single-factor experimental
results. The optimal combination of independent variables in oil extraction based on the
extracted oil yield (Y1 ) and PV (Y2 ) as response variables was acquired using RSM. To
check the variability of the RSM experiment, the central points were repeated three times.
All experiments were carried out twice and the full second-order polynomial regression
model was developed as follows:

3 3 3
Y = β0 + ∑ βi Xi + ∑ βii X2i + ∑ βij Xi Xj (4)
i=1 i=1 i6 =j=1

where Y represents the response variable, β0 is the regression intercept, and βi , βii , and βij
are the regression coefficients for the linear, quadratic, and interactive effects of the model,
respectively. Xi and Xj are the independent variables [31].
An analysis of variance (ANOVA) was performed to determine the lack of fit and
the effect of the linear, quadratic, and interaction terms of each of the response variables.
The analysis of the experimental data and the optimization process were carried out using
Minitab Statistical Software Cloud-Based Version (Minitab LLC., State College, PA, USA).
Differences between the means were analyzed based on a significance level of p < 0.05,
using IBM SPSS Statistics Version 28.0 (Thaisoftup Co., Ltd., Bangkok, Thailand).
Foods 2022, 11, 291 5 of 16

3. Results and Discussion

3.1. Chemical Composition of Silkworm Pupae
The nutrient compositions of the silkworm pupae are presented in Table 1. In fresh
samples, silkworm pupae had a protein content of approximately 11.7%, which is higher
than the 10.3% protein content reported for the edible insect Acheta domesticus [32], but not
as high as the 16% and 24% protein content reported for the eri silkworm pupae grown on
tapioca and castor leaves, respectively [33], and the 24.4% protein content of another edible
insect, Euxoa auxiliaries [34]. Based on the definition of FAO/WHO, the majority of insects
and pupae are regarded as high in protein (over 10 g proteins/100 g edible portion of
fresh weight basis) [35]. A fat content of around 6.1% was observed in the silkworm pupae
used in this study, which was lower than that found in eri silkworm pupae (8.0–8.6%) and
reported by Longvah et al. [33]. Furthermore, the oil content of the silkworm Bombyx mori in
the pupal stage was different between fresh male (4.8%) and female (9.0%) pupae [10]. Total
carbohydrate and dietary fiber (~4.0% and 1.1%, respectively) comparable to the values
reported by Longvah et al. were also found in the silkworm pupae [33]. The silkworm
pupae had an energy of approximately 521 kcal/100 g dried pupae, which is in the range
of the caloric values for insects and pupae (293–762 kcal/100 g dry weight) as reported by
Rumpold and Schlüter [36]. These values depend on the insects’ diet, species, and life stage.
Two major nutrients, protein and fat (~51.9% and 26.9%, respectively), were consistent with
what was reported in previous works [7,33]. In general, insects and pupae are composites
containing proteins, lipids, and polysaccharides. Silkworm pupae have been reported to be
30–55% protein, 10–15% carbohydrates, and 25–30% oil, which is rich in various saturated
and unsaturated fatty acids [37–41]. The high protein and fat content of silkworm pupae
indicate their extensive potential as an alternative source for food and feed.

Table 1. The nutrient composition of silkworm pupae.

Content (per 100 g)

Nutrient
Fresh Pupae Dry Basis
Moisture (g) 77.43 ± 0.08 -
Energy (kcal) 117.55 ± 0.49 520.69 ± 3.96
Protein (g) 11.72 ± 0.01 51.89 ± 0.15
Total fat (g) 6.07 ± 0.15 26.87 ± 0.75
Total carbohydrate (g) 4.03 ± 0.21 17.83 ± 0.85
Dietary fiber (g) 1.07 ± 0.06 4.72 ± 0.30
Ash (g) 0.77 ± 0.01 3.41 ± 0.05

3.2. Effect of a Single Factor on the Extraction Yield

3.2.1. Effect of Stirring Time
Figure 1a shows the effect of stirring time on the extraction yield. The results indicate
a significant increase in the extraction yield after increasing the stirring time from 60 min
(2.35% ± 0.05%) to 120 min (3.35% ± 0.11%) and a slight decrease with an additional
increase in stirring time to 180 min (2.79% ± 0.40%) (p ≤ 0.05). The reason for this could be
that a longer stirring time leads to the sample’s emulsification, which may then trap the oil
in the system and result in a decline in the oil’s yield [19,42,43]. Therefore, a longer stirring
time is unnecessary once the maximum extraction yield is achieved. Based on the results,
stirring time should be controlled in the range of 120 min, which is sufficient to attain the
extraction yield for this treatment.
 aggregation and precipitation. This phenomenon will result in the trapping of oil by pro-
tein, which will restrict the release of oil and lower the yield of extracted oil [43,45,46].
Hashemi et al. [47] reported that the addition of NaCl also decreased the extraction time
and reduced the required temperature for the extraction of Vitex pseudonegundo by ohmic-
assisted hydrodistillation. However, when increasing the saline solution above 2% w/v,
Foods 2022, 11, 291 6 of 16
the extraction yields significantly decreased (p ≤ 0.05), and it should be noted that protein
denaturation affected the oil extraction [48].

Figure
Figure 1.
1. The
The effect
effect of
of stirring
stirring time
time (a),
(a), NaCl
NaCl concentration
concentration (b),
(b), and
and liquid-to-solid
liquid-to-solid ratio
ratio (c)
(c) on
on the
the
extraction
extraction yield
yield of
of silkworm
silkworm pupae
pupae oil.
oil. Different
Different letters
letters indicate
indicate significant differences (p
significant differences (p ≤
≤ 0.05).
0.05).

3.2.2. Effect
3.2.3. Effect of
of the
the Liquid-to-Solid
Concentration of NaCl
Ratio
The silkworm
The pupaeratio
liquid-to-solid oil was
is anextracted
importantwith various
factor thatconcentrations
can influence of theNaCl by an aque-
extraction effi-
ous extraction method. As shown in Figure 1b, the extraction yields
ciency. Different liquid-to-solid ratios (1–5 mL/g) were determined, and the results positively correlated
are
with increasing
presented concentrations
in Figure of NaCl
1c. It was found from
that the0% to 2% w/v
extraction and
yield reached
tended to the highest
increase withvalue
in-
creasing ± 0.11%
of 3.35% liquid when
saline using 2%
volume, w/v NaCl.
beginning withThis value
2.37% significantly
± 0.13% decreased when
at a liquid-to-solid ratiousing
of 1
NaCl and
mL/g above this concentration
reaching a maximum ≤ 0.05). This
(pextraction could
yield be because
of 3.35% ± 0.11% inorganic salt increases
at a liquid-to-solid the
ratio
polarity difference between oil and water, thereby reducing the oil’s solubility
of 3 mL/g. After this point, the extraction yields moderately decreased and remained con- in the water
and enhancing
stant (~2.70%) withthe process’s salt-assisted
a continuous increaseeffect
in the[44].
ratioInfrom
other4 to
words,
5 mL/ga higher concentration
(p ≤ 0.05). This may
be because the oil–water separation effect was minimized with a high ratioprecipitation.
of NaCl can cause a salting-out effect that leads to protein aggregation and of water to
This phenomenon
material [44]. The will result
results in the trapping
suggested of oil by amount
that a suitable protein, which willsolution
of saline restrict the
canrelease
easily
of oil and
achieve thelower the yield
maximum of extracted
extraction yieldoil
of[43,45,46].
the target Hashemi
compounds. et al.Although
[47] reported
a largethatamount
the ad-
dition
of of NaCl
extraction also decreased
solvent can increasethe extraction time and
the leaching-out reduced
rates thecompounds,
of target required temperature for
it also leads
the extraction of Vitex pseudonegundo by ohmic-assisted hydrodistillation.
to the waste of extraction solvent and an extra layer of complexity in the extraction proce- However, when
[49]. the saline solution above 2% w/v, the extraction yields significantly decreased
increasing
dure
(p ≤ 0.05), and it should be noted that protein denaturation affected the oil extraction [48].

3.2.3. Effect of the Liquid-to-Solid Ratio

The liquid-to-solid ratio is an important factor that can influence the extraction effi-
ciency. Different liquid-to-solid ratios (1–5 mL/g) were determined, and the results are
presented in Figure 1c. It was found that the extraction yield tended to increase with
increasing liquid saline volume, beginning with 2.37% ± 0.13% at a liquid-to-solid ratio
of 1 mL/g and reaching a maximum extraction yield of 3.35% ± 0.11% at a liquid-to-solid
ratio of 3 mL/g. After this point, the extraction yields moderately decreased and remained
constant (~2.70%) with a continuous increase in the ratio from 4 to 5 mL/g (p ≤ 0.05).
This may be because the oil–water separation effect was minimized with a high ratio of
water to material [44]. The results suggested that a suitable amount of saline solution can
easily achieve the maximum extraction yield of the target compounds. Although a large
amount of extraction solvent can increase the leaching-out rates of target compounds, it also
Foods 2022, 11, 291 7 of 16

leads to the waste of extraction solvent and an extra layer of complexity in the extraction
procedure [49].

3.3. Effect of Independent Processing Parameters on Response Variables from the BBD Experiment
The effect of different levels of the independent variables, including the NaCl con-
centration, liquid-to-solid ratio, and stirring time, on the yield (Y1 ), PV (Y2 ), and AV (Y3 )
of silkworm pupae oil was determined at different treatment combinations, and all the
results are summarized in Table 2. The yield of extracted silkworm pupae oil varied in the
range of 2.39–3.50 wt%, with a mean value of ~2.82 wt%, or accounting for ~54.7% recovery
oil (Equation (S1)). The differences in the fat content of the silkworm pupae oil varied
according to its species, origin, and metamorphosis stage and the extraction’s method
and conditions [36,50–52]. For example, Shanker et al. [52] studied the characteristics
of the neutral lipid of desilked eri silkworm pupae (Samia cynthia ricini) fed with castor
(Ricinus communis Linn.) and tapioca (Manihot utilissima Pohl.) leaves. The oil content in
the pupae was in the range of 18–20% (dry basis). In addition, Longvah et al. [33] reported
that the fat content observed in eri silkworm prepupae grown on either castor or tapioca
was not different (~26.2%), while the eri silkworm pupae grown on caster had higher fat
content than those grown on tapioca (25%) (dry basis). The fat content tended to be higher
during the stages of pupae, larvae, and wintering than during other periods for the same
species [51].
The PV and AV are important quality parameters related to the extent of the oil’s
oxidation and hydrolysis, respectively. The statistical analysis of the results showed that
the PV of extracted oil was significantly dependent on the extraction conditions (p ≤ 0.05).
However, the observed values varied from 1.40 to 2.10 mM (Table 2) and fell within the
suggested value of ~6.69 mM or 15 meq/kg oil (Table S1) for virgin oils [53]. The AV is
responsible for the release of free fatty acids. In this study, the AV of oil varied from 0.53 to
0.75 mg KOH/g oil (Table 2). These values are lower than the maximum recommended
value of 4 mg KOH/g for virgin fats and oils [53]. However, there were no significant
(p > 0.05) effects of the extraction conditions on the AV of extracted oils. This means that
the NaCl concentration, liquid-to-solid ratio, and stirring time did not affect the acidity of
the silkworm pupae oil.

3.4. Model Fitting

A multiple regression analysis was performed based on the experimental data (Table 2),
and the backward second-order polynomial regression models as a function of independent
variables were established for the extracted yield (Y1 ) and PV (Y2 ) of the silkworm pupae
oil. The estimated coefficients of the independent variables are shown in Table 3, and the
equations in terms of actual levels are presented as follows:

Y1 = −5.650 + 2.186X1 + 1.691X2 + 0.070X3 − 0.567X21 − 0.249X22 − 0.0003X23 (5)

Y2 = 4.048 + 0.088X1 − 1.620X2 + 0.247X22 (6)

The adequacy and predictability of the models were considered according to the
determination coefficient (R2 ), adjusted determination coefficient (R2 adj ), lack of fit, and
absolute average deviation (AAD) [31]. For the extracted oil yield, the R2 and R2 adj were
0.948 and 0.908, respectively (Table 3), suggesting the model’s excellent fitting degree,
i.e., it could interpret more than 90% of the variation in the response variables studied.
Considering the PV results in which the R2 and R2 adj were 0.704 and 0.623, respectively, we
observed a good fitting degree of the model, which could interpret more than 62% of the
variation in the response variables studied [54]. To support the adequacy and how well
the model fitted the data, a lack-of-fit test was used. In this work, p-values of the model’s
“lack of fit” were 0.480 and 0.802 for the extracted yield and PV of the silkworm pupae oil,
respectively. These values suggested that the “lack of fit” had no statistical significance
(p > 0.05), which verified the reliability of the models [55].
Foods 2022, 11, 291 8 of 16

Table 2. The experimental run and extraction conditions for the Box–Behnken design (BBD) with
coded levels of each variable (blanket), including the response values for the extraction yield, PV, and
AV of silkworm pupae oil.

NaCl Liquid-to- Extraction Yield PV

Stirring Time (%, Y1 ) (mM, Y2 ) AV
Run Concentration Solid Ratio (min, X3 ) (mg KOH/g, Y3 )
(% w/v, X1 ) (mL/g, X2 ) Exp Pred Exp Pred
1 2 (0) 4 (1) 90 (−1) 3.25 ± 0.02 c 3.10 1.64 ± 0.03 cd 1.70 0.61 ± 0.09 a
2 2 (0) 4 (1) 150 (1) 3.09 ± 0.05 d 2.98 1.64 ± 0.05 cd 1.70 0.56 ± 0.14 a
3 3 (1) 3 (0) 90 (−1) 2.51 ± 0.02 gh 2.51 1.66 ± 0.03 cd 1.68 0.72 ± 0.10 a
4 1 (−1) 4 (1) 120 (0) 2.78 ± 0.01 e 2.83 1.72 ± 0.19 cd 1.61 0.64 ± 0.15 a
5 3 (1) 4 (1) 120 (0) 2.57 ± 0.03 fg 2.67 1.82 ± 0.14 bc 1.78 0.75 ± 0.18 a
6 3 (1) 3 (0) 150 (1) 2.48 ± 0.04 gh 2.33 1.74 ± 0.12 c 1.68 0.71 ± 0.11 a
7 1 (−1) 3 (0) 150 (1) 2.60 ± 0.06 fg 2.55 1.40 ± 0.03 e 1.50 0.64 ± 0.15 a
8 1 (−1) 3 (0) 90 (−1) 2.65 ± 0.01 f 2.67 1.72 ± 0.24 cd 1.50 0.53 ± 0.15 a
9 1 (−1) 2 (−1) 120 (0) 2.57 ± 0.03 fg 2.44 1.74 ± 0.15 c 1.88 0.64 ± 0.15 a
10 2 (0) 3 (0) 120 (0) 3.30 ± 0.04 bc 3.37 1.41 ± 0.03 e 1.59 0.68 ± 0.08 a
11 2 (0) 2 (−1) 90 (−1) 2.62 ± 0.11 fg 2.71 1.99 ± 0.05 ab 1.97 0.64 ± 0.15 a
12 2 (0) 3 (0) 120 (0) 3.50 ± 0.09 a 3.37 1.73 ± 0.08 cd 1.59 0.72 ± 0.10 a
13 2 (0) 2 (−1) 150 (1) 2.55 ± 0.08 gh 2.59 2.10 ± 0.10 a 1.97 0.55 ± 0.06 a
14 3 (1) 2 (−1) 120 (0) 2.39 ± 0.02 h 2.27 2.06 ± 0.10 a 2.06 0.63 ± 0.09 a
15 2 (0) 3 (0) 120 (0) 3.38 ± 0.04 ab 3.37 1.48 ± 0.22 de 1.59 0.62 ± 0.08 a
Mean values followed by different letters in the same column indicate significant differences (p ≤ 0.05). Exp and
Pred represent experimental and predicted values, respectively.

Table 3. The estimated coefficient of the independent variables with the associated statistical signifi-
cance of each coefficient for backward regression models.

Regression Extraction Yield (%, Y1 ) PV (mM, Y2 )

Terms a Coefficient F-Value p-Value b Coefficient F-Value p-Value b
Model 24.17 0.000 8.71 0.003
Constant
−5.650 4.048
(β0 )
X1 (β1 ) 2.186 4.04 0.079 0.088 3.71 0.080
X2 (β2 ) 1.691 23.57 0.001 1.620 8.75 0.013
X3 (β3 ) 0.070 0.93 0.363
X1 X1 (β11 ) −0.567 91.80 0.000
X2 X2 (β22 ) −0.249 17.76 0.003 0.247 13.66 0.004
X3 X3 (β33 ) −0.0003 20.30 0.002
R2 0.948 0.704
Adjusted
0.908 0.623
R2
Lack of fit 1.37 0.480 0.51 0.802
AAD (%) 2.760 5.581
a X1 , X2 , and X3 represent the NaCl concentration (% w/v), liquid-to-solid ratio (mL/g), and stirring time (min),
respectively. b p-value more than 0.05 is not significantly different at a 5% level.

Furthermore, AAD, another realized parameter, was also calculated to measure the
models’ adequacy and accuracy [31]. The suggested range of 0% to 30% for AAD was
reported in a previous work by Bas and Boyaci [56]. For the present study, the AAD values
obtained for the oil yield and PV were small, with values of 2.76% and 5.58%, respectively
(Table 3). These small values confirmed the satisfactory fit of the regression models to the
experimental data. This was consistent with the correlation plots of the experimental or
actual and predicted values (Figure 2). A linear distribution was observed for both the oil
yield and PV data, indicating well-fitting models (R2 ~0.999 and 0.996, respectively). The
normal probability plot is also presented in Figure 3 to show the normality of the internally
studentized residual. The plot indicated that the residuals (the difference between the
actual and predicted values) followed a normal distribution and formed an approximately
straight line, and the fitted model provided a reasonable estimate for oil extraction yields.
According to the criteria discussed above, these results indicated that the regression mod-
 XX33 (β (β33)) 0.070
0.070 0.93
0.93 0.363
0.363
XX11X X11 (β(β1111)) −0.567
−0.567 91.80
91.80 0.000
0.000
XX22X X22 (β(β2222)) −0.249
−0.249 17.76
17.76 0.003
0.003 0.247
0.247 13.66
13.66 0.004
0.004
XX33X X33 (β(β3333)) −0.0003
−0.0003 20.30
20.30 0.002
0.002
RR22 0.948
0.948 0.704
0.704
Foods 2022, 11, 291 9 of 16
Adjusted
Adjusted R R 2
2 0.908
0.908 0.623
0.623
Lack of
Lack of fit
fit 1.37
1.37 0.480
0.480 0.51
0.51 0.802
0.802
AAD (%)
AAD (%) 2.760
2.760 5.581
5.581
aa els
X are
X11,, X
X adaptable
22,, and
and X andthe
X33 represent
represent can
the be concentration
NaCl
NaCl used as response
concentration surface
(% w/v),
(% w/v), models ratio
liquid-to-solid
liquid-to-solid for an
ratio estimation
(mL/g),
(mL/g), of the
and stirring
and stirring
mean
time
time (min),
(min), response.
respectively. bb p-value
respectively. p-value more
more than
than 0.05
0.05 is
is not
not significantly
significantly different
different atat aa 5%
5% level.
level.

Figure2.
Figure
Figure Thecorrelation
2.2.The
The correlationplots
correlation plotsof
plots ofofactual
actualand
actual andpredicted
and predictedvalues
predicted valuesof
values ofofthe
theextraction
the extractionyield
extraction yield(a)
yield (a)and
(a) andperoxide
and peroxide
peroxide
value (b).
value(b).
value (b).

Figure3.
Figure
Figure 3.3. The
The
The plots
plots
plots of of normal
of normal
normal probability
probability
probability of of residuals
of residuals
residuals for the
for for the extraction
the extraction
extraction yield (a)
yield yield
(a) and
and (a) and
peroxide
peroxide peroxide
value
value (b).
(b).
value (b).
3.5. Analysis
3.5. Analysis of of the
the Response
Response Surface
Surface
3.5. Analysis of the Response Surface
The interaction
The interaction between
between the the response
response and and the
the independent
independent variables
variables was
was visually
visually
The interaction between the response and the independent variables was visually
interpreted using a response surface and contour plots. The plots represented
interpreted using a response surface and contour plots. The plots represented the response the response
interpreted using a response surface and contour plots. The plots represented the response
level against
level against two independent
independent variables at at a central level
level of the remaining
remaining independent
level againsttwo two independentvariables variables ataacentral
central levelofofthethe remainingindependent
independent
variable. The
variable. The response
response surface
surface and
and contour
contour plots
plots ofof the
the extracted
extracted yield
yield of
of silkworm pupae pupae
variable. The response surface and contour plots of the extracted yield ofsilkworm silkworm pupae
oil are
oil are shown
shown in in Figure
Figure 44 for
for the
the interaction
interaction between
between the the NaCl
NaCl concentration
concentration and and liquid-
liquid-
oil are shown in Figure 4 for the interaction between the NaCl concentration and liquid-to-
to-solid
to-solid ratio
ratio
solid ratio (Figure
(Figure
(Figure 4a,b),
4a,b),
4a,b), NaCl
NaCl
NaCl concentration
concentration
concentration and
and
and stirring
stirring
stirring time
time
time (Figure
(Figure
(Figure 4c,d),and
4c,d),
4c,d), andliquid-to-
and liquid-
liquid-
to-solid
to-solid ratio
ratio
solid ratio and and
and stirring
stirring
stirring time
time
time (Figure
(Figure
(Figure 4e,f).
4e,f).
4e,f). Similarly,
Similarly,
Similarly, these these
these plots
plotsplots between
between
between NaCl
NaClNaCl concen-
concen-
concentration
tration
tration and
and liquid-to-solid
liquid-to-solid ratio
ratio for
for the
the PV
PV of
of extracted
extracted oil
oil are
are also
also shown
shown
and liquid-to-solid ratio for the PV of extracted oil are also shown in Figure 5. Figure 4a,b in
in Figure
Figure 5.
5.
indicate the interaction effect between the NaCl concentration and the liquid-to-solid ratio
on the oil yield extracted at a 120 min stirring time. The oil yields gradually increased
with an increase in NaCl concentration from 1 to 2% w/v at the designated liquid-to-solid
ratio and then decreased with an increase in the concentration of NaCl. An increase in the
concentration of NaCl has advantages for extraction because an increase in the polarity of
this solution leads to a decrease in the solubility of nonpolar compounds, thus resulting
in a high extracted yield [46]. The disadvantage of using excess NaCl concentrations is a
salting-out effect, which promotes the aggregation and precipitation of proteins and affects
the extraction process [57]. Similarly, the oil yields obviously increased and then decreased
with the increase in the liquid-to-solid ratio at any NaCl concentration. These results are
probably due to the decrease in the oil–water separation effect because of the high ratio of
water to material [45].
 of NaCl concentration. An increase in the oil yield was seen when the stirring time was
increased from 90 to 120 min. However, the oil yield tended to decrease with longer stir-
ring times from 120 to 150 min. This effect was observed for all levels of liquid-to-solid
ratios used in this study. The increase in oil yields with increasing stirring times was
mainly ascribed to the fact that a longer stirring time could facilitate the oil transfer from
Foods 2022, 11, 291 10 of 16
the ruptured pupae tissue particles to the aqueous phase. This trend was also observed in
a recent study, where the extraction time had a positive effect on the oil yield [45,58].

Foods 2022, 11, x FOR PEER REVIEW 11 of 16

NaCl concentration at any liquid-to-solid ratio. The results indicated that the PV of silk-
worm pupae oil was greater at both low and high levels of liquid-to-solid ratios at the
designated NaCl concentration. At an intermediate level of the liquid-to-solid ratio, how-
Figure
ever, the4. Response surface (a,c,e)
PV of extracted and contour
oil displayed theplots (b,d,f)
lowest value function
as a(less
functionthanof1.5 independent
independent
mM). Thevariables
increase
on
in the
PVextraction yield. (a,b)
with an increase inNaCl
NaClconcentration
concentration andmay
liquid-to-solid
be because ratio, (c,d)promotes
NaCl NaCl concentration
peroxide
and
and stirring
stirring time,
formation, time, (e,f)
leading stirring
(e,f)tostirringtime
timeand
an increase inliquid-to-solid
and liquid-to-solid
peroxide ratio. X1,XX1presence
ratio.
values. The 2,, X
and X3 represent
2 , and X NaClthe
of3 representNaCl con-
the NaCl
is known to
centration,
concentration,liquid-to-solid
liquid-to-solidratio, and
ratio, stirring
and time,
stirring respectively.
time, respectively.
increase oxidation since it can readily donate its valence electron [59,60].
Regarding the PV, the response surface and contour plots (Figure 5a,b, respectively)
as a function of the NaCl concentration and the liquid-to-solid ratio at a 120 min stirring
time were established according to the significant terms in the regression model; see Equa-
tion (6). It showed that the PV of silkworm pupae oil was positively influenced by the

Figure5.5.The
Figure Theresponse
responsesurface
surface(a)(a)and
andcontour plots
contour (b)(b)
plots as aasfunction of independent
a function of independentvariables on the
variables on
the XPV.
PV. X1 and
1 and X2 represent
X2 represent the NaCl
the NaCl concentration
concentration and and liquid-to-solid
liquid-to-solid ratio,ratio, respectively.
respectively.

Figure 4c,d represent

3.6. Optimization the interaction
and Validation effect between
of the Regression Model the NaCl concentration and stirring
time on the oil yield extracted at a liquid-to-solid ratio of 3 mL/g. An increase in the stirring
Response optimization was performed to analyze the optimal levels of independent
time improved the oil yields to some degree at any NaCl concentration, but a further
variables required to achieve the maximum extraction yield and minimum PV of silk-
increase in the stirring time caused a decrease in oil yields. A similar effect in the stirring
worm pupae oil. To determine the exact levels for all the independent variables necessary
time as shown in Figure 4e,f was also observed, indicating an interaction effect between
to achieve the optimal conditions, a numerical optimization was utilized. Results showed
the liquid-to-solid ratio and the stirring time on the oil yield extracted at 2% w/v of NaCl
that the extraction using a 1.707% w/v NaCl solution at a 3.273 mL/g liquid-to-solid ratio
concentration. An increase in the oil yield was seen when the stirring time was increased
and a 119.307 min stirring time provided the optimal maximum extraction yield and min-
from 90 to 120 min. However, the oil yield tended to decrease with longer stirring times
imum PV with a desirability of 1000 (Table 4). After considering the feasibility of these
from 120 to 150 min. This effect was observed for all levels of liquid-to-solid ratios used in
parameters during real production, the optimum conditions were modified to an NaCl
this study. The increase in oil yields with increasing stirring times was mainly ascribed to
solution
the of 1.7%
fact that w/v,stirring
a longer a liquid-to-solid
time could ratio of 3.3
facilitate themL/g, and a from
oil transfer stirring
the time of 119
ruptured min.
pupae
Under these conditions, the predicted maximum yield and minimum PV of
tissue particles to the aqueous phase. This trend was also observed in a recent study, where silkworm pu-
pae oil were 3.38% and 1.54 mM, respectively. Using these
the extraction time had a positive effect on the oil yield [45,58]. same conditions, the average
values of the extraction
Regarding yield
the PV, the and PVsurface
response of silkworm pupae oil
and contour obtained
plots (Figure from the
5a,b, experiment
respectively)
were
as 3.32% ±of0.03%
a function and concentration
the NaCl 1.55 ± 0.02 mM, andrespectively. These experimental
the liquid-to-solid ratio at a 120results corre-
min stirring
time were established according to the significant terms in the regression model; (p
sponded fairly well to the predicted values, which were not significantly different see>
0.05) from the predicted values (Table 4). The relative error values between the predicted
and actual oil yield and PV were also calculated as 1.81% and 0.65%, respectively. This
revealed that the models were well fitted for the extraction of oil from silkworm pupae
under optimal aqueous saline conditions, and the designed models were good for predict-
ing the optimal extraction conditions.
Foods 2022, 11, 291 11 of 16

Equation (6). It showed that the PV of silkworm pupae oil was positively influenced by
the NaCl concentration at any liquid-to-solid ratio. The results indicated that the PV of
silkworm pupae oil was greater at both low and high levels of liquid-to-solid ratios at
the designated NaCl concentration. At an intermediate level of the liquid-to-solid ratio,
however, the PV of extracted oil displayed the lowest value (less than 1.5 mM). The increase
in PV with an increase in NaCl concentration may be because NaCl promotes peroxide
formation, leading to an increase in peroxide values. The presence of NaCl is known to
increase oxidation since it can readily donate its valence electron [59,60].

3.6. Optimization and Validation of the Regression Model

Response optimization was performed to analyze the optimal levels of independent
variables required to achieve the maximum extraction yield and minimum PV of silkworm
pupae oil. To determine the exact levels for all the independent variables necessary to
achieve the optimal conditions, a numerical optimization was utilized. Results showed that
the extraction using a 1.707% w/v NaCl solution at a 3.273 mL/g liquid-to-solid ratio and a
119.307 min stirring time provided the optimal maximum extraction yield and minimum
PV with a desirability of 1000 (Table 4). After considering the feasibility of these parameters
during real production, the optimum conditions were modified to an NaCl solution of
1.7% w/v, a liquid-to-solid ratio of 3.3 mL/g, and a stirring time of 119 min. Under these
conditions, the predicted maximum yield and minimum PV of silkworm pupae oil were
3.38% and 1.54 mM, respectively. Using these same conditions, the average values of
the extraction yield and PV of silkworm pupae oil obtained from the experiment were
3.32% ± 0.03% and 1.55 ± 0.02 mM, respectively. These experimental results corresponded
fairly well to the predicted values, which were not significantly different (p > 0.05) from
the predicted values (Table 4). The relative error values between the predicted and actual
oil yield and PV were also calculated as 1.81% and 0.65%, respectively. This revealed that
the models were well fitted for the extraction of oil from silkworm pupae under optimal
aqueous saline conditions, and the designed models were good for predicting the optimal
extraction conditions.

Table 4. A comparison between the experimental and predicted values for the response variables at
optimal conditions of extraction yield and PV of silkworm pupae oil.

Response Variables Experimental Value Predicted Value p-Value a

Extraction yield
3.32 ± 0.03 3.38 0.154
(wt%)
PV (mM) 1.55 ± 0.02 1.54 0.147
a p-value more than 0.05 is not significantly different at a 5% level (one-sample t-test).

3.7. Chemical Compositions of Silkworm Pupae Oil Extracted by Aqueous Saline

3.7.1. Fatty Acid Composition
The fatty acid composition of the silkworm pupae oil obtained from aqueous saline
extraction is presented in Table 5. The most abundant fatty acid in the extracted silkworm
pupae oil was oleic acid (C18:1, 36.84 ± 0.03), which generally exists as a monounsaturated
fatty acid (MUFA) in most edible fats and oils. Silkworm pupae oil is a source of essential
polyunsaturated fatty acids (PUFAs) named α-linolenic (C18:3, 24.85 ± 0.06) and linoleic
acid (C18:2, 4.25 ± 0.01), which belong to n-3 (omega-3) and n-6 (omega-6) series fatty acids,
respectively. The total saturated fatty acids found in the extracted silkworm pupae oil
accounted for approximately 33% of the total fatty acids. Palmitic (C16:0, 26.0% ± 0.05%)
and stearic acids (C18:0, 6.78 ± 0.01) were the main acids found in the extracted oil. The
composition of fatty acid observed in the silkworm pupae oil corresponded with previously
published data [7,8]. However, it is important to note that the higher and lower ratios in
the fatty acid composition are due to differences in extraction methods and factors, species’
origin, sex, maturation stage, season, and geographical regions, as reported in previous
studies [2,9].
Foods 2022, 11, 291 12 of 16

Table 5. The fatty acid composition, content of minor compounds, and DPPH free radical scavenging
activity of silkworm pupae oil extracted by aqueous saline methods.

Criteria Content
Fatty acid composition (% Total fatty acids)
Saturated fatty acids 33.10 ± 0.05
Caprylic acid (C8:0) 0.12 ± 0.01
Lauric acid (C12:0) 0.05 ± 0.00
Myristic acid (C14:0) 0.16 ± 0.00
Palmitic acid (C16:0) 26.0 ± 0.05
Stearic acid (C18:0) 6.78 ± 0.01
Monounsaturated fatty acids (MUFA) 37.82 ± 0.03
Palmitoleic acid (C16:1) 0.98 ± 0.00
Oleic acid (C18:1) 36.84 ± 0.03
Polyunsaturated fatty acids (PUFA) 29.09 ± 0.07
Linoleic acid (18:2, n-6) 4.25 ± 0.01
Linolenic acid (18:3, n-3) 24.85 ± 0.06
Minor compounds
Cholesterol (mg/100 g) 108.66 ± 0.09
β-Carotene (µg/100 g) 784.89 ± 12.17
α-Tocopherol (µg/100 g) 9434.39 ± 367.95
DPPH free radical scavenging activity (%) 57.59 ± 0.93

3.7.2. Minor Components

As shown in Table 5, the cholesterol level in the extracted silkworm pupae oil was
108.7 mg/100 g, which is comparable to beef tallow [61,62] and slightly higher than is seen
in lard and chicken fat [61]. However, the cholesterol level of the silkworm pupae oil was
lower than those found in fish oils and milk fat [62–64]. The cholesterol content in extracted
silkworm pupae oil was lower than what was reported by Belluco et al. and Winitchai
et al. [65,66]; it has been shown that B. mori chrysalises contain a high concentration of
cholesterol (214 mg/100 g) [60]. In addition, it was found that the average cholesterol con-
tent in four types of insects’ lipids including termites’ (Macrotermes bellicosus), caterpillars’
(Imbrasia belina), and beetle larvae of Oryctes rhinoceros and Rhynchophorus phoenicis were
around 3.6% [67].
The contents of naturally occurring antioxidants, especially carotenoids and toco-
pherols, are the most important factor affecting the oxidative stability of edible oils. There-
fore, the amount of β-carotene and α-tocopherol in the extracted silkworm pupae oil was
determined and found to be approximately 785 and 9434 µg/100 g oil, respectively. Re-
garding the content of β-carotene, it was found to be lower than the content of crude palm
oil [68] but higher than the content of crude rapeseed and sunflower oils [69], cold-pressed
hemp oil, flax oil, and canola seed oil [70]. Regarding α-tocopherol, the silkworm pupae oil
extracted by the aqueous saline method in our study contained less α-tocopherol than some
crude and refined vegetable oils [71,72]. However, it was higher than the values reported
for cold-pressed hemp and flax seed oils [70]. Higher and lower values of the α-tocopherol
content of silkworm pupae oil had been reported in previous studies and are agreed to be
due to differences in sex, maturation stage, and extraction method [11,66,73].

3.8. DPPH Free Radical Scavenging Activity

Lipid oxidation accelerates when free radicals form as a result of unsaturated fatty
acids losing hydrogen atoms from their double bonds [74]. The scavenging of free rad-
icals initiates the oxidation mechanism and inhibits the chain reaction, thus preventing
lipid oxidation [75]. The DPPH free radical scavenging effect of silkworm pupae oil was
57.59 ± 0.93%. This result was comparable to a previous study revealing the antioxidant ac-
tivity of the muga silkworm (Antheraea assamensis) [76] and suggested that silkworm pupae
oil had a noticeable effect on scavenging DPPH free radicals. In general, silkworm pupae
oil contains tocopherols, natural phenolic antioxidants that are good sources of vitamin E
Foods 2022, 11, 291 13 of 16

and play an important role as lipid oxidation inhibitors [11]. The results demonstrated that
the extracted silkworm pupae oil had good quality parameters, making it a useable and
valuable source of edible lipids.

4. Conclusions
This study identified that aqueous saline extraction is an efficient, green, and simple
method for silkworm oil extraction. An RSM-based BBD was successfully used to evaluate
the influence of independent variables, including the NaCl concentration, aqueous liquid to
silkworm ratio, and extraction time, on the oil’s yield and quality. The optimal conditions of
the extraction process were found to be 1.7% w/v NaCl, a 3.3 mL/g liquid-to-solid ratio, and
a 119 min extraction time. Under these extraction conditions, the highest yield (3.32 wt%)
and highest-quality oil (PV and AV of 1.55 mM and 0.67 mg KOH/g oil, respectively) were
obtained. Overall, the analyzed results of the fatty acid composition and the content of
minor components demonstrated that silkworm oil obtained from this alternative extraction
method has the potential to be used as a source of edible oil in the food industry. The
comparison of the extraction efficiency, quality, composition, and nutritional values of this
oil with other extraction methods such as mechanical, solvent extraction, and supercritical
fluid will be carried out in a separate study.

Supplementary Materials: The following are available online at https://www.mdpi.com/article/10.3

390/foods11030291/s1, Equation (S1), Table S1. Peroxide values (PV) of silkworm pupae oil in mM
and meq/kg oil.
Author Contributions: Conceptualization, M.P., W.K., T.H. and U.K.; methodology, J.T., C.P.T. and
U.K.; formal analysis, J.T. and U.K.; investigation, J.T.; resources, U.K.; data curation, J.T. and U.K.;
writing—original draft preparation, J.T., C.P.T. and U.K.; writing—review and editing, J.T., M.P., W.K.,
T.H., C.P.T. and U.K.; visualization, J.T. and U.K.; supervision, M.P., W.K., T.H. and U.K.; project
administration, U.K.; funding acquisition, U.K. All authors have read and agreed to the published
version of the manuscript.
Funding: This research was funded by the Kasetsart University Research and Development In-
stitute (KURDI) under the project Development of Advance Researcher Competence System for
Competitiveness in Agriculture and Food, grant number FF(KU) 25.64.
Data Availability Statement: The datasets generated for this study are available on request to the
corresponding author.
Acknowledgments: The authors acknowledge the use of laboratory facilities and research equipment
provided by the Department of Food Science and Technology and the Faculty of Agro-Industry,
Kasetsart University. We also thank Prakit Sukyai from the Department of Biotechnology, Faculty of
Agro-Industry, Kasetsart University for providing the centrifugation instrument.
Conflicts of Interest: The authors declare no conflict of interest.

References
1. Manjunath, R.N.; Kumar, A.; Kumar, K.P.A. Utilisation of Sericulture Waste by Employing Possible Approaches. In Contaminants
in Agriculture Sources, Impacts and Management; Naeem, M., Ansari, A.A., Gill, S.S., Eds.; Springer: Cham, Switzerland, 2020;
pp. 385–399.
2. Hu, B.; Li, C.; Zhang, Z.; Zhao, Q.; Zhu, Y.; Su, Z.; Chen, Y. Microwave-assisted extraction of silkworm pupal oil and eval-
uation of its fatty acid composition, physicochemical properties and antioxidant activities. Food Chem. 2017, 231, 348–355.
[CrossRef] [PubMed]
3. Sonjui, T.; Noomhorm, C.; Promboon, A. Sericin recovery from silk cocoon degumming wastewater by a membrane process.
Kasetsart J. (Nat. Sci.) 2009, 43, 538–549.
4. Rao, P.; Rathod, V. Valorization of Food and Agricultural Waste: A Step towards Greener Future. Chem Rec. 2019,
19, 1858–1871. [CrossRef]
5. Hirunyophat, P.; Chalermchaiwat, P.; On-nom, N.; Prinyawiwatkul, W. Selected nutritional quality and physicochemical
properties of silkworm pupae (frozen or powdered) from two species. Int. J. Food Sci. 2021, 56, 3578–3587. [CrossRef]
6. Longvah, T.; Manghtya, K.; Qadri, S.S. Eri silkworm: A source of edible oil with a high content of α-linolenic acid and of
significant nutritional value. J. Sci. Food Agric. 2012, 92, 1988–1993. [CrossRef] [PubMed]
Foods 2022, 11, 291 14 of 16

7. Pereira, N.R.; Ferrarese-Filho, O.; Matsushita, M.; de Souza, N.E. Proximate composition and fatty acid profile of Bombyx mori L.
chrysalis toast. J. Food Compost Anal. 2003, 16, 451–457. [CrossRef]
8. Rao, P.U. Chemical composition and nutritional evaluation of spent silk worm pupae. J. Agric. Food Chem. 1994,
42, 2201–2203. [CrossRef]
9. Ray, M.; Gangopadhyay, D. Effect of maturation stage and sex on proximate, fatty acid and mineral composition of eri silkworm
(Samia ricini) from India. J. Food Compost. Anal. 2021, 100, 103898. [CrossRef]
10. Saviane, A.; Tassoni, L.; Naviglio, D.; Lupi, D.; Savoldelli, S.; Bianchi, G.; Cortellino, G.; Bondioli, P.; Folegatti, L.; Casartelli, M.
Mechanical processing of Hermetia illucens larvae and Bombyx mori pupae produces oils with antimicrobial activity. Animals 2021,
11, 783. [CrossRef] [PubMed]
11. Kotake-Nara, E.; Yamamoto, K.; Nozawa, M.; Miyashita, K.; Murakami, T. Lipid profiles and oxidative stability of silkworm
pupal oil. J. Oleo Sci. 2002, 51, 681–690. [CrossRef]
12. Pan, W.J.; Liao, A.M.; Zhang, J.G.; Dong, Z.; Wei, Z.J. Supercritical carbon dioxide extraction of the oak silkworm (Antheraea
pernyi) pupal oil: Process optimization and composition determination. Int. J. Mol. Sci. 2012, 13, 2354–2367. [CrossRef]
13. Srinivas, G.; Nidoni, U.; Ramachandra, C.T.; Ramappa, K.T.; Ashoka, J. Supercritical fluid extraction of pupae oil from mulberry
silkworm (Bombyx mori L.). J. Pharmacogn Phytochem. 2019, 8, 4507–4513.
14. Wei, Z.-J.; Liao, A.-M.; Zhang, H.-X.; Liu, J.; Jiang, S.-T. Optimization of supercritical carbon dioxide extraction of silkworm pupal
oil applying the response surface methodology. Bioresour. Technol. 2009, 100, 4214–4219. [CrossRef]
15. Aitta, E.; Marsol-Vall, A.; Damerau, A.; Yang, B. Enzyme-assisted extraction of fish oil from whole fish and by-products of Baltic
herring (Clupea harengus membras). Foods 2021, 10, 1811. [CrossRef]
16. Li, X.; Cao, J.; Bai, X.; Zhang, F. Chemical composition and thermal properties of tilapia oil extracted by different methods. Int. J.
Food Prop. 2018, 21, 1575–1585. [CrossRef]
17. Okada, T.; Morrissey, M.T. Recovery and characterization of sardine oil extracted by pH adjustment. J. Agric. Food Chem. 2007,
55, 1808–1813. [CrossRef] [PubMed]
18. Ramakrishnan, V.V.; Ghaly, A.E.; Brooks, M.S.; Budge, S.M. Extraction of oil from mackerel fish processing waste using alcalase
enzyme. Enzym. Eng. 2013, 2, 1–10.
19. Tzompa-Sosa, D.A.; Yi, L.; van Valenberg, H.J.F.; Lakemond, C.M.M. Four insect oils as food ingredient: Physical and chemical
characterisation of insect oils obtained by an aqueous oil extraction. J. Insects Food Feed. 2019, 5, 279–292. [CrossRef]
20. Saha, B.C.; Hayashi, K. Debittering of protein hydrolyzates. Biotechnol. Adv. 2001, 19, 355–370. [CrossRef]
21. Yoon, S.; Wong, N.A.K.; Chae, M.; Auh, J.-H. Comparative characterization of protein hydrolysates from three edible insects:
Mealworm larvae, adult crickets, and silkworm pupae. Foods 2019, 8, 563. [CrossRef]
22. Sui, Y.; Huang, W.-C.; Wu, Y.; Qi, X.; Mao, X. Lipid extraction from Greenland halibut (Reinhardtius hippoglossoides) by-product in
low-voltage DC electric field and its mechanism. J. Clean. Prod. 2021, 283, 124673. [CrossRef]
23. Cruz, S.; Yousfi, K.; Pérez, A.G.; Mariscal, C.; Garcia, J.M. Salt improves physical extraction of olive oil. Eur. Food Res. Technol.
2007, 225, 359–365. [CrossRef]
24. Kadioglu, S.I.; Phan, T.T.; Sabatini, D.A. Surfactant-based oil extraction of corn germ. J. Am. Oil Chem. Soc. 2011,
88, 863–869. [CrossRef]
25. Prichapan, N.; McClements, D.; Klinkesorn, U. Encapsulation of iron within W1 /O/W2 emulsions formulated using a
natural hydrophilic surfactant (saponin): Impact of surfactant level and oil phase crystallization. Food Biophysics. 2020,
15, 346–354. [CrossRef]
26. AOCS. Official and Recommended Practices of the AOCS; AOCS Press Publication: Champaign, IL, USA, 1997.
27. AOAC. Official Methods of Analysis of AOAC International, 21st ed.; AOAC International: Rockville, MD, USA, 2019.
28. Speek, A.J.; Schrijver, J.; Schreurs, W.H.P. Vitamin E composition of some seed oils as determined by high-performance liquid
chromatography with fluorometric detection. J. Food Sci. 1985, 50, 121–124. [CrossRef]
29. Speek, A.J.; Temalilwa, C.R.; Schrijver, J. Determination of β-carotene content and vitamin A activity of vegetables by high-
performance liquid chromatography and spectrophotometry. Food Chem. 1986, 19, 65–74. [CrossRef]
30. Singh, R.P.; Chidambara Murthy, K.N.; Jayaprakasha, G.K. Studies on the antioxidant activity of pomegranate (Punica granatum)
peel and seed extracts using in vitro models. J. Agric. Food Chem. 2002, 50, 81–86. [CrossRef] [PubMed]
31. Yolmeh, M.; Jafari, S.M. Applications of response surface methodology in the food industry processes. Food Bioprocess Technol.
2017, 10, 413–433. [CrossRef]
32. Pennino, M.; Dierenfeld, E.; Behler, J. Retinol, α-tocopherol and proximate nutrient composition of invertebrates used as feed. Int.
Zoo Yearbook. 2007, 30, 143–149. [CrossRef]
33. Longvah, T.; Mangthya, K.; Ramulu, P. Nutrient composition and protein quality evaluation of eri silkworm (Samia ricinii)
prepupae and pupae. Food Chem. 2011, 128, 400–403. [CrossRef]
34. White, D.; Kendall, K. Grizzly bears and army cutworm moths in the alpine of Glacier National Park, Montana. Int. Bear News.
1993, 2, 2–3.
35. FAO; WHO. Food Labelling; FAO: Rome, Italy, 2007.
36. Rumpold, B.A.; Schlüter, O.K. Nutritional composition and safety aspects of edible insects. Mol. Nutr. Food Res. 2013,
57, 802–823. [CrossRef]
Foods 2022, 11, 291 15 of 16

37. Zhang, R.; Rao, Z.; Li, Y.; Li, H.; Fei, L.; Lei, S.; Wang, Y. Silkworm excrement derived in-situ co-doped nanoporous carbon as
confining sulfur host for lithium sulfur batteries. Chem. Select. 2019, 4, 5678–5685. [CrossRef]
38. Xu, X.-H.; Liu, Z.-X.; Shi, X.-Y.; Miao, C.; Sheng, S.; Xu, Y.; Wu, F.-A.; Wang, J. Fed-batch fermentation of Yarrowia Lipolytica using
defatted silkworm pupae hydrolysate: A dynamic model-based approach for high yield of lipid production. Waste Biomass Valoriz.
2018, 9, 2399–2411. [CrossRef]
39. Wang, W.; Wang, N.; Zhou, Y.; Zhang, Y.; Xu, L.; Xu, J.; Feng, F.; He, G. Isolation of a novel peptide from silkworm pu-
pae protein components and interaction characteristics to angiotensin I-converting enzyme. Eur. Food Res. Technol. 2011,
232, 29–38. [CrossRef]
40. Bhavsar, P.; Fontana, G.D.; Tonin, C.; Patrucco, A.; Zoccola, M. Superheated water hydrolyses of waste silkworm pupae protein
hydrolysate: A novel application for natural dyeing of silk fabric. Dyes Pigm. 2020, 183, 108678. [CrossRef]
41. Tomotake, H.; Katagiri, M.; Yamato, M. Silkworm pupae (Bombyx mori) are new sources of high quality protein and lipid. J. Nutr.
Sci. Vitaminol. (Tokyo) 2010, 56, 446–448. [CrossRef]
42. Matsakidou, A.; Mantzouridou, F.T.; Kiosseoglou, V. Optimization of water extraction of naturally emulsified oil from maize
germ. LWT-Food Sci. Tech. 2015, 63, 206–213. [CrossRef]
43. Rosenthal, A.; Pyle, D.L.; Niranjan, K. Aqueous and enzymatic processes for edible oil extraction. Enzyme Microb. Technol. 1996,
19, 402–420. [CrossRef]
44. Huang, Y.; Yin, Z.; Guo, J.; Wang, F.; Zhang, J. Oil extraction and evaluation from yellow horn using a microwave-assisted
aqueous saline process. Molecules 2019, 24, 2598. [CrossRef] [PubMed]
45. Maseer Nejres, A. Salting Out. 2020. Available online: https://www.researchgate.net/publication/347936827_salting_out
(accessed on 3 December 2021).
46. Yi, L.; Van Boekel, M.A.J.S.; Lakemond, C.M.M. Extracting Tenebrio molitor protein while preventing browning: Effect of pH and
NaCl on protein yield. J. Insects as Food Feed. 2017, 3, 21–31. [CrossRef]
47. Hashemi, S.M.B.; Kamani, M.H.; Amani, H.; Mousavi Khaneghah, A. Voltage and NaCl concentration on extraction of essential
oil from Vitex pseudonegundo using ohmic-hydrodistillation. Ind. Crops Prod. 2019, 141, 111734. [CrossRef]
48. Li, W.; Zhou, R.; Mu, Y. Salting effects on protein components in aqueous NaCl and urea solutions: Toward understanding of
urea-induced protein denaturation. J. Phys. Chem. B 2012, 116, 1446–1451. [CrossRef]
49. Zhang, L.; Wang, M. Optimization of deep eutectic solvent-based ultrasound-assisted extraction of polysaccharides from Dioscorea
opposita Thunb. Int. J. Biol. Macromol. 2017, 95, 675–681. [CrossRef]
50. Williams, J.P.; Williams, J.R.; Kirabo, A.; Chester, D.; Peterson, M. Nutrient content and health benefits of insects. In Insects as
Sustainable Food Ingredients; Elsevier Inc.: Amsterdam, The Netherlands, 2016; pp. 61–84.
51. Zou, Y.-X.; Hu, T.-G.; Shi, Y.; Liu, J.; Mu, L.-X.; Xiao, Y.; Liao, S.-T. Establishment of a model to evaluate the nutritional quality of
Bombyx mori Linnaeus (Lepidoptera, Bombycidae) pupae lipid based on principal components. J. Asia Pac. Entomol. 2017, 20,
1364–1371. [CrossRef]
52. Shanker, K.S.; Shireesha, K.; Kanjilal, S.; Kumar, S.V.; Srinivas, C.; Rao, J.V.; Prasad, R.B. Isolation and characterization of neutral
lipids of desilked eri silkworm pupae grown on castor and tapioca leaves. J. Agric. Food Chem. 2006, 54, 3305–3309. [CrossRef]
53. Alimentarius, C. Codex Standard for edible fats and oils not covered by individual standards. Codex Stan 1981, 19, 1–6.
54. Granato, D.; de Araújo Calado, V.M.; Jarvis, B. Observations on the use of statistical methods in food science and technology. Food
Res. Int. 2014, 55, 137–149. [CrossRef]
55. Hippolyte, M.; Augustin, M.; Tchougang, H.; Robert, N.; Somashekar, D. Application of response surface methodology to improve
the production of antimicrobial biosurfactants by Lactobacillus paracasei subsp. tolerans N2 using sugar cane molasses as substrate.
Bioresour. Bioprocess. 2018, 5, 1–16. [CrossRef]
56. Baş, D.; Boyacı, İ.H. Modeling and optimization I: Usability of response surface methodology. J. Food Eng. 2007,
78, 836–845. [CrossRef]
57. Jiang, Y.; Zhu, Y.; Zheng, Y.; Liu, Z.; Zhong, Y.; Deng, Y.; Zhao, Y. Effects of salting-in/out-assisted extractions on structural, physic-
ochemical and functional properties of Tenebrio molitor larvae protein isolates. Food Chem. 2021, 338, 128158. [CrossRef] [PubMed]
58. Tian, Y.; Xu, Z.; Zheng, B.; Lo, Y.M. Optimization of ultrasonic-assisted extraction of pomegranate (Punica granatum L.) seed oil.
Ultrason. Sonochem. 2013, 20, 202–208. [CrossRef]
59. Zhao, B.; Zhou, H.; Zhang, S.; Pan, X.; Li, S.; Zhu, N.; Wu, Q.; Wang, S.; Qiao, X.; Chen, W. Changes of protein oxidation, lipid
oxidation and lipolysis in Chinese dry sausage with different sodium chloride curing salt content. Food Sci. Hum. Wellness. 2020,
9, 328–337. [CrossRef]
60. Mi, E. Effect of some preservatives on the peroxide value and carbohydrate content of dry melon seeds. J. Chem. Pharm.
2017, 9, 139–142.
61. Farmani, J.; Rostammiri, L. Characterization of chicken waste fat for application in food technology. J. Food Meas. Charact. 2015, 9,
143–150. [CrossRef]
62. Sabir, S.M.; Hayat, I.; Gardezi, S.D.A. Estimation of sterols in edible fats and oils. Pak. J. Nutr. 2003, 2, 178–181.
63. Chee, K.M.; Gong, J.X.; Rees, D.M.G.; Meydanl, M.; Ausman, L.; Johnson, J.; Siguel, E.N.; Schaefer, E.J. Fatty acid content of
marine oil capsules. Lipids 1990, 25, 523–528. [CrossRef]
64. Precht, D. Cholesterol content in European bovine milk fats. Die Nahrung. 2001, 45, 2–8. [CrossRef]
Foods 2022, 11, 291 16 of 16

65. Belluco, S.; Losasso, C.; Maggioletti, M.; Alonzi, C.C.; Paoletti, M.G.; Ricci, A. Edible insects in a food safety and nutritional
perspective: A critical review. Compr. Rev. Food Sci. Food Saf. 2013, 12, 296–313. [CrossRef]
66. Winitchai, S.; Manosroi, J.; Abe, M.; Boonpisuttinant, K.; Manosroi, A. Free radical scavenging activity, tyrosinase inhibition
activity and fatty acids composition of oils from pupae of native Thai silkworm (Bombyx mori L.). Agric. Nat. Resour. 2011,
45, 404–412.
67. Ekpo, K.; Onigbinde, A.; Asia, I.O. Pharmaceutical potentials of the oils of some popular insects consumed in southern Nigeria.
Afr. J. Pharm. Pharmacol. 2009, 3, 51–57.
68. Nagendran, B.; Unnithan, U.; Choo, Y.; Sundram, K. Characteristics of red palm oil, a carotene-and vitamin E–rich refined oil for
food uses. Food Nutr. Bull. 2000, 21, 189–194. [CrossRef]
69. Kreps, F.; Vrbiková, L.; Schmidt, Š. Influence of industrial physical refining on tocopherol, chlorophyll and beta-carotene content
in sunflower and rapeseed oil. Eur. J. Lipid Sci. Technol. 2014, 116, 1572–1582. [CrossRef]
70. Teh, S.-S.; Birch, E. Physicochemical and quality characteristics of cold-pressed hemp, flax and canola seed oils. J. Food Compost
Anal. 2013, 30, 26–31. [CrossRef]
71. Grilo, E.C.; Costa, P.N.; Gurgel, C.S.S.; Beserra, A.F.D.L.; Almeida, F.N.D.S.; Dimenstein, R. Alpha-tocopherol and gamma-
tocopherol concentration in vegetable oils. Food Sci. Technol. 2014, 34, 379–385. [CrossRef]
72. Vidrih, R.; Vidakovič, S.; Abramovič, H. Biochemical parameters and oxidative resistance to thermal treatment of refined and
unrefined vegetable edible oils. Czech J. Food Sci. 2010, 28, 376–384. [CrossRef]
73. Ravinder, T.; Kaki, S.S.; Kunduru, K.R.; Kanjilal, S.; Rao, B.; Swain, S.; Prasad, R.B.N. Physico-chemical characterization and
oxidative stability studies of eri silkworm oils. Int. J. Modern Chem. Appl. Sci. 2016, 3, 293–300.
74. Samaram, S.; Mirhosseini, H.; Tan, C.P.; Ghazali, H.M.; Bordbar, S.; Serjouie, A. Optimisation of ultrasound-assisted extraction
of oil from papaya seed by response surface methodology: Oil recovery, radical scavenging antioxidant activity, and oxidation
stability. Food Chem. 2015, 172, 7–17. [CrossRef] [PubMed]
75. Liu, C.T.; Wu, C.Y.; Weng, Y.M.; Tseng, C.Y. Ultrasound-assisted extraction methodology as a tool to improve the antioxidant
properties of herbal drug Xiao-chia-hu-tang. J. Ethnopharmacol. 2005, 99, 293–300. [CrossRef]
76. Deori, M.; Boruah, D.C.; Devi, D.; Devi, R. Antioxidant and antigenotoxic effects of pupae of the muga silkworm Antheraea
assamensis. Food Biosci. 2014, 5, 108–114. [CrossRef]