Available via license: CC BY-NC-ND 4.0
Content may be subject to copyright.
Changes in nutritional compositions and digestive enzyme inhibitions of
isoavone-enriched soybean leaves at different stages (drying, steaming,
and fermentation) of food processing
Hee Yul Lee
a,1
, Ji Ho Lee
a,1
, Du Yong Cho
a
, Kyeong Jin Jang
a
, Jong Bin Jeong
a
, Min Ju Kim
a,b
,
Ga Young Lee
a
, Mu Yeun Jang
a
, Jin Hwan Lee
c,*
, Kye Man Cho
a,*
a
Department of GreenBio Science and Agri-Food Bio Convergence Institute, Gyeongsang National University, Naedong-ro 139-8, Jinju 52849, Republic of Korea
b
Biological Resources Utilization Division, National Institute of Biological Resources (NIBR), Sangnam-ro 1008-11, Miryang, 50452, Republic of Korea
c
Department of Life Resource Industry, Dong-A University, 37, Nakdong-daero 550 beon-gil, Saha-gu, Busan 49315, Republic of Korea
ARTICLE INFO
Keywords:
Isoavone-enriched soybean leaves
Food processing
γ-Aminobutyric acid
Water-soluble vitamin
Isoavone
Digestive enzyme inhibitory
ABSTRACT
Isoavone-enriched soybean leaves (IESLs) were processed for drying, steaming, and fermentation, and bioactive
compounds and biological activities were analyzed. During food processing, the content of fatty acids, water-
soluble vitamins, total phenolics, total avonoids, and isoavone-aglycones increased from dried IESLs
(DrIESLs) to fermented IESLs (FeIESLs). Especially, oleic acid (53.4 → 113.1 mg/100 g, 2.1-folds), γ-amino-
butyric acid (357.36 → 435.48 mg/100 g, 1.2-folds), niacin (19.0 → 130.6 mg/100 g, 6.9-folds), folic acid (9.7
→ 25.5 mg/100 g, 2.6-folds), daidzein (270.02 → 3735.10
μ
g/100 g, 13.8-folds), and genistein (121.18 →
1386.01
μ
g/100 g, 11.4-folds) dramatically increased. Correspondingly, the antioxidant and digestive enzyme
inhibitory activities increased. Therefore, solid-state lactic acid fermentation (SLAF) was suggested as a suitable
technique for mass-processing IESLs. FeIESLs with SLAF have the potential to be utilized as a functional food.
1. Introduction
In the contemporary food market, soybeans come in various types
and have become a globally popular food known for its versatility and
functional benets. However, in this situation, we have overlooked
soybean leaves. Unlike the common soybeans and soybean-based
products that we commonly encounter in our routine lives, the utiliza-
tion of soybean leaves is rare, mostly because of the lack of knowledge
about soybean leaves. Among the key substances well known to be
abundant in soybeans are vegetable proteins and functional substances
such as isoavones. Interestingly, isoavones are also present in soybean
leaves (Lee et al., 2023; Yuk et al., 2016). It was especially reported that
soybean leaves treated with ethephon in the eld have more isoavone
content compared to normal soybean leaves and were named isoavone-
enriched soybean leaves (IESLs) because of these results. In the case of
IESLs, isoavones exist that glycosides (daidzin, genistin, malo-
nyldaidzin, and malonylgenistin) and aglycones (daidzein and genis-
tein)(Lee et al., 2023; Xie et al., 2021; Yuk et al., 2016). However, there
are limits to obtaining isoavones by consuming soybean leaves as a
general food. A processing method is needed to effectively obtain iso-
avones from IESLs. Additionally, for commercial use, it is important to
identify processing methods suitable for mass production (Yuk et al.,
2016).
Current trends in the global isoavone market include a focus on the
extraction, analysis methods, and biological activities of these com-
pounds (Campos, 2021). Especially the numerous biological attributes
of soy isoavone were reported in several studies. Isoavone was re-
ported also known as phytoestrogens, are recognized as benecial
components for middle-aged woman owing to their potential to relieve
symptoms associated with menopause (Hirose et al., 2016; Lee et al.,
2017). Additionally, research has expanded to support their preventa-
tive effects on adult diseases such as anticancer, osteoporosis, and heart
diseases (Basu & Maier, 2018; Xie et al., 2021). These trends highlight
the diverse applications and potential of isoavones in various in-
dustries, driving innovation and research in this eld.
Fermentation is extremely important and widespread in our daily
lives, and each country has its representative fermented foods. A di-
versity of microorganisms, including fungi, yeast, and lactic acid
* Corresponding authors.
E-mail addresses: schem72@dau.ac.kr (J.H. Lee), kmcho@gnu.ac.kr (K.M. Cho).
1
These authors contributed equally to this work.
Contents lists available at ScienceDirect
Food Chemistry: X
journal homepage: www.sciencedirect.com/journal/food-chemistry-x
https://doi.org/10.1016/j.fochx.2024.101999
Received 14 August 2024; Received in revised form 10 November 2024; Accepted 11 November 2024
Food Chemistry: X 24 (2024) 101999
Available online 14 November 2024
2590-1575/© 2024 The Authors. Published by Elsevier Ltd. This is an open access article under the CC BY-NC-ND license (
http://creativecommons.org/licenses/by-
nc-nd/4.0/ ).
bacteria (LAB), are used for fermentation. According to previous reports,
LAB fermentation can convert to fatty acids, amino acids, and iso-
avones in fermentation substrates, which from linolenic acid, glutamic
acid, and glycoside isoavones to conjugated linoleic acid, γ-amino-
butyric acid (GABA), and aglycone isoavones (Lee et al., 2018; Lee
et al., 2022). These converted compounds have health benets such as
anti-obesity, reduced blood pressure, and anti-diabetes (Chen et al.,
2018; Lee et al., 2018; Lee et al., 2022). Especially, aglycones thus
produced are easily digestible, easily absorbed in the body, and have
higher bioavailability than glycosides (Kim et al., 2022; Liu et al., 2023;
Lu et al., 2024). Also, the reported increase of functional components
was higher in combination fermentation than in single LAB (Lee et al.,
2018).
Therefore, in this study, optimal mass processing methods were
conrmed by comparing dried IESLs (DrIESLs), steamed IESLs (StIESLs),
and fermented IESLs (FeIESLs). At that time, two LAB strains with
excellent probiotic activity, Lactiplantibacillus plantarum P1201 and
Levilactobacillus brevis BMK184, were used for solid-state lactic acid
fermentation (SLAF) of IESLs. For this, we investigated nutritional
compositions and digestive enzyme inhibitions.
2. Materials and methods
2.1. Preparation of IESLs and starters
The IESLs were provided in the dried state by JC & Farm Corporation
(Namhae-gun, Gyeongsangnam-do, Republic of Korea), and they were
named DrIESLs. At that time, the IESLs were prepared with the following
methods. First, the soybean leaves were grown until the growth stage R3
(the beginning of pod development) at a time with a growth period of
about 60 days after planting, and the height of the plant was 120 cm on
average. Then, an ethephon of 200
μ
g/mL concentration was sprayed on
soybean leaves of the R3 growth stage until the solution dripped every
24 h and 2 times. That is, the ethephon solution (200
μ
g/mL) was treated
at an amount of 30 ±5 mL per plant. Finally, the treated soybean leaves
were harvested and called IESLs (Yuk et al., 2016). Provided DrIESLs
were ground and stored at −40 ◦C in a freezer (MDF
–
U5412, Panasonic
Co., Osaka, Japan); they were used as needed. Highly active probiotic
strains L. plantarum P1201 and L. brevis BMK184 were used for
fermentation. These microorganisms were previously isolated from the
traditional Kimchi and bitter melon fermentation (Lee et al., 2018; Lee
et al., 2022).
2.2. Culture medium, chemical regents, and instruments
The microbial culture medium and chemical reagents for analysis of
viable cell numbers, isoavones, and biological activities (including
antioxidant and digestive enzyme inhibitory activities) were used in the
same manner as previously described by our research team. The same
instruments, such as gas chromatography (GC, Nexis GC-2030, Shimad-
zu Corp., Kyoto, Japan), automatic amino acid analyzer (L-8900, Hitachi
High-Technologies Co., Tokyo, Japan), high-performance liquid chro-
matography (HPLC, Agilent 1200 system, Agilent Technologies Inc.,
Waldbronn, Germany), and spectrophotometer (UV-1800 240 V, Shi-
madzu Corp., Kyoto, Japan) were used for measuring the content of fatty
acids, amino acids, isoavones, total phenolics (TP), and total avonoids
(TF), as described previously (Lee et al., 2022; Lee et al., 2023).
2.3. Processing methods of IESLs
Processing of IESLs was performed as Lee et al. (2022) reports. Before
this study was performed, we had already studied optimizing fermen-
tation time for IESLs with L. plantarum P1201 and L. brevis BMK184.
IESLs were fermented at 25, 30, and 35 ◦C, respectively. As a result of
that, the optimal fermentation time was 35 ◦C. For the fermentation
process, DrIESLs were mixed with sucrose (2 % of DrIESLs weight) and
tap water (4-fold of DrIESLs weight, v/w) and sterilized at 121 ◦C for 30
min; it was named StIESLs. And then cooled to prepare the FeIESLs, the
starter cultures (with a mixed ratio of L. plantarum P1201: L. brevis
BMK184 =1:1) at 5 % (v/v) were inoculated and then fermented at
35 ◦C for 72 h. Then, StIESLs and FeIESLs were collected, dried (for 48 h
at 55 ◦C), and ground (100 mesh size) for use in the experiments.
2.4. Determination of physicochemical properties, viable cell numbers,
and proximate compositions
The pH measurement was performed by using a pH meter (Orion
STAR A211, Thermo Fisher Scientic, Waltham, MA, USA), and 1 g of
the sample was mixed with 50 mL of distilled water via neutralization
titration, and the neutralized titration was performed to pH 8.2 ±0.1
with 0.1 N NaOH. Thereafter, an appropriate amount of 0.1 N NaOH was
calculated and converted into lactic acid equivalents. The viable cell
counts were determined by serial dilution using sterile water, followed
by plating on MRS agar plates and incubating at 37 ◦C for 48 h for colony
enumeration. The analysis of proximate compositions was conducted in
accordance with AOAC (Latimer, 2012): The moisture content was
measured using the air oven method, while ash content was determined
with a mufe furnace. Protein and fat content were analyzed through
the Kjeldahl (using Kjeldahl apparatus, DNP-1500, Raypa, Seoul, Korea)
and the Soxhlet (using Soxtec, ST243, Foss Analytical Co., Luoyang,
China) method, respectively. Carbohydrate content was calculated using
the difference method. These results were presented as percentages of
dry weight (DW).
2.5. Preparation of extracts
The extracts were prepared by modifying the method described
previously by Lee et al. (2023). The extraction was performed by adding
40-fold 80 % ethanol to 1 g of each powder sample and allowing it to
stand at room temperature (20 ±5 ◦C) for 12 h. The extract was then
centrifuged for 30 min, and the supernatant was ltered through a 0.45-
μ
m membrane lter (Dismic-25CS, Toyoroshikaisha, Ltd., Tokyo,
Japan). This ltered extract was used for the determination of TP, TF,
and isoavone contents. For the measurements of the biological activ-
ities, ltered extracts were evaporated using a rotary evaporator (N-
1300, SHANGHAI EYELA Co., Ltd., Shanghai, China) and freeze-dried
after being maintained at −40 ◦C.
2.6. Determination of fatty acid contents
For fatty acids analysis, the sample preparation and preprocessing
were performed using the modied method of Lee et al. (2018). Briey,
the course was performed as per the fatty acid esterication process.
First, 25 mg of sample powder, 0.5 mL of 0.5 N NaOH dissolved in
methanol, and 0.5 mL of triundecanoin as an internal standard (C
11:0
, 2
mg/mL), were mixed with vortex. The mixture heated at 100 ◦C for 5
min. After, 2 mL of 14 % BF
3
(Supelco, Bellefonte, PA, USA) was mixed,
and heated at 100 ◦C for 30 min. Then, 1 mL of isooctane was added and
vigorously stirred for 30 s. Immediately, add 5 mL of saturated NaCl
solution and shake. After cooling to 25 ◦C, separate the isooctane layer
from the aqueous layer and dehydrate it with sodium sulte anhydrous.
The pretreated samples were ltered with a 0.45
μ
m-membrane for
analysis with GC. Finally, the fatty acid amount of each sample was
analyzed using a GC equipped with an SP-2560 capillary column (100 m
×0.25 mm i.d., 0.2
μ
m lm thickness, Supelco, St. Louis, MO, USA) and
a ame ionization detector. The injection volume and the injector
temperature were set at 1
μ
L and 225 ◦C, operating in split mode with a
split ratio of 200:1. The carrier gas (Helium) was own at a rate of 0.75
mL/min. The initial oven temperature was held at 100 ◦C for 4 min, then
increased to 240 ◦C at a rate of 3 ◦C/min, and held at 240 ◦C for 15 min.
H.Y. Lee et al.
Food Chemistry: X 24 (2024) 101999
2
2.7. Determination of free amino acid contents
To analyze the free amino acids, the sample preparation and pre-
processing were performed in accordance with the amino acid analysis
method a previously described method by Khan et al. (2018). Briey, the
program was prosecuted according to the amino acid derivatization
process. The treated samples were centrifuged at 15,000 ×g for 3 min
and ltered through a syringe lter. The ltrates were concentrated by
using a rotary evaporator (N-1300, Tokyo Rikakikai Co., Ltd., Tokyo,
Japan) at 50 ◦C. Subsequently, 2 mL of lithium buffer (pH 2.2) was
added to dissolve the concentrate, and the solution was ltered through
a 0.45-
μ
m membrane lter. The nal ltered sample was quantitatively
analyzed by using an automatic amino acid analyzer.
2.8. Determination of water-soluble vitamin contents
The analysis of water-soluble vitamins was conducted in accordance
with the vitamin analysis a slight modication of the method described
by Lee et al. (2023). First, 10 mL of 50 % MeOH was added to 1 g of each
sample, followed by extraction for 12 h, and centrifugation at 4000 ×g to
separate the supernatant. The supernatant was then ltered through a
0.45-
μ
m membrane lter. Subsequently, the analysis was performed
using HPLC. The analysis was conducted by using the LiChrospher® Si
60 column (LichroCART 250–4, 5
μ
m, 125 mm ×4 mm, Merck KGaA,
Darmstadt, Germany). Mobile phase solvent A consisted of HPLC water
composed of 0.1 % folic acid, while mobile phase solvent B was
composed of HPLC acetonitrile. The analysis conditions included a
sample injection volume of 20
μ
L, a ow rate of 1 mL/min, a column
temperature of 30 ◦C, and a detection wavelength of 256 nm. The linear
gradient for solvent B was as follows: 0 % (0–5 min), 0–75 % (5–15 min),
and 75 % (15–25 min).
2.9. Determination of isoavones
The analysis of isoavones was conducted using HPLC following the
method described by Lee et al. (2022). The analysis was performed using
Lichrospher 100 RP C18 columns (LichroCART 125–4, 5
μ
m, 125 mm ×
4 mm, Merck KGaA, Darmstadt, Germany) as the analysis column.
Mobile phase solvent A was composed of HPLC water containing 0.2 %
acetic acid, while mobile phase solvent B was composed of HPLC
acetonitrile containing 0.2 % acetic acid. The gradient for the mobile
phase was set based on the mobile phase solvent B as follows: 0 % (0
min), 10 % (15 min), 20 % (25 min), 25 % (35 min), and 35 % (45–50
min). The analysis conditions included a sample injection volume of 20
μ
L, a ow rate of 1 mL/min, a column temperature of 30 ◦C, and a
detection wavelength of 254 nm using a diode array detector. The ex-
periments were conducted in pentaplicate and the HPLC chromatograms
of the six isoavone derivatives were detected. The isoavones in the
sample were identied by matching the retention time of respective
standards and then quantied using linear calibration curves for each
standard, applying external standardization. The calibration curves
were developed using seven concentration points (1.0, 0.8, 0.6, 0.4, 0.2,
0.1, and 0.05 mg/mL) derived from a stock solution with a concentra-
tion of 1 mg/mL for each standard, correlation coefcient (r
2
) higher
than 0.998.
2.10. Determination of TP and TF contents
TP and TF contents were determined according to the Folin-Denis
and Dvis’s methods, respectively, with Lee et al. (2022). The TP and
TF contents were calculated with a standard curve using gallic acid and
rutin (concentrations: 1, 5, 10, 20, 50, 100, 200, and 500
μ
g/mL),
respectively.
2.11. Determination of antioxidant activities
For determining the antioxidant activities, such as 2,2-diphenyl-1-
picrylhydrazyl (DPPH), 2,2
′
-azinobis-(3-ethylbenzothiazoline-6-sul-
fonic acid) (ABTS), hydroxyl radical activities, and ferric reducing/
antioxidant power (FRAP), the reactions were performed according to a
previously described method by Zhang et al. (2015) and Lee et al.
(2023). The three radical scavenging activities were determined by
measuring the absorbance at 525, 732, and 520 nm, respectively. In all
cases, the absorbance of the test sample and the negative control were
measured, and the percentage (%) was calculated as follows (1). FRAP
measurements were performed by determining the absorbance at 525
nm.
Radical scavenging activity (%) = [1−(Asample/Acontrol ) ] ×100 (1)
Asample =absorbance in sample,Acontrol =absorbance of control.
2.12. Determination of digestive enzyme inhibitory activities
To determine the digestive enzyme inhibitory activities, including
α
-glucosidase and pancreatic lipase inhibitory activities, the reactions
were performed according to the previously described method by Zhang
et al. (2015) and Lee et al. (2023). The reactions were then terminated
by adding 750
μ
L of 100 mM Na
2
CO
3
, and the absorbance was measured
at 420 nm, respectively. For the negative control, the sample solvent was
used instead of the sample, and the absorbance of both the experimental
and negative control groups was measured and expressed as a percent-
age (%), calculated by the following formula (2).
Digestive enzyme inhibitory activity (%)
=[1−(Asample/Acontrol ) ] ×100 (2)
Asample =absorbance of sample,Acontrol =absorbance of control.
2.13. Statistical analysis and data processing
All experiments were performed in pentaplicate, and the results were
expressed as the mean values. The results of all experiments were pre-
sented as mean ±standard deviation using analysis of variance
(ANOVA) conducted with the Statistical Analysis System (SAS 9.4, SAS
Institute, Cary, NC, USA). For assessing the signicance of the ANOVA
results, Duncan’s multiple range tests was performed at a signicance
level of p <0.05. We performed principal component analysis (PCA) and
clustering heatmap analyses in R software version 4.3.3 (R Project for
Statistical Computing, Vienna, Austria).
3. Results and discussion
3.1. Comparison of physicochemical properties, viable cell counts, and
proximate compositions in the DrIESLs, StIESLs, and FeIESLs
The analysis results of physicochemical properties and proximate
composition are shown in supplementary Table S1. In terms of the pH,
acidity, and viable cell numbers, the FeIESLs indicated lactic acid
fermentation. The pH decreased from 6.19 to 4.89, and the acidity
increased from 0.39 to 1.10 % in an inverse proportion to pH. The viable
cell numbers measured 10.77 log CFU/g at the fermentation sample
(FeIESLs). In the proximate composition analysis, the moisture rate
slightly increased (11.4 → 12.2 and 11.9 %), and the protein rate was
not signicantly different from 30.5 to 29.0 %. The carbohydrate rate
slightly increased (45.0 → 46.9 and 46.0 %), and the ash and fat contents
signicantly increased from 6.9 and 5.2 % to 7.3 and 5.8 %, respectively.
LAB primarily utilizes carbohydrates as carbon sources and small
peptides and amino acids as nitrogen sources, thereby exhibiting varied
auxotrophism depending on the species and strains (Khan et al., 2018).
H.Y. Lee et al.
Food Chemistry: X 24 (2024) 101999
3
Consequently, the substances generated during fermentation can vary
with the species, strains, and environmental conditions of the LAB. Thus,
selecting appropriate substrates, environmental conditions, and strains
is crucial for obtaining the desired compounds (Okabe et al., 2011). In
our study, based on the literature data, IESLs were cocktail-fermented
using L. plantarum P1201 and L. brevis BMK184 and then categorized
into DrIESLs, StIESLs, and FeIESLs (Lee et al., 2018; Lee et al., 2022).
Our fermentation process was effectively conducted, as indicated by the
analysis of physicochemical properties. A decrease in pH and an increase
in the titratable acidity observed were attributed to the accumulation of
organic acids produced through lactic acid fermentation. Furthermore,
the concurrent increase in viable cell counts and a decrease in the pH
value can be explained by the proliferation of LAB. Both in the
fermentation of soy milk using L. plantarum WCFS1 and in the single and
mixed fermentation of soybean leaves using L. plantarum P1201 and
L. brevis BMK184, the pH value decreased along with an increase in the
viable cell count, thereby exhibiting different changes depending on the
fermentation strains and duration (Lee et al., 2022; Papadia et al., 2018).
In addition, the decrease in pH can induce acid stress, potentially
leading to various chemical transformations. Thus, the acidic environ-
ment resulting from lactic acid fermentation may serve as a crucial
factor (Saubade et al., 2017). The mass FeIESLs were shown a similar
pattern of small scare fermentation soybean leaves in previous research.
3.2. Comparison of fatty acid contents in the DrIESLs, StIESLs, and
FeIESLs
The results of fatty acid analysis according to the processing stages
are shown in Table 1and Fig. 1. The relatively pronounced changes were
observed in γ-linolenic acid, docosadienoic acid, eicosenoic acid, nerv-
onic acid, arachidic acid, and behenic acid. These components showed a
low overall proportion, and they were not detected in FeIESLs. Oleic
acid, the major unsaturated fatty acid, exhibited an increasing trend
from 53.4 to 113.1 mg/100 g, while linoleic acid exhibited a slight in-
crease after a decrease from 144.5 to 136.9 mg/100 g. In the case of
saturated fatty acids, palmitic acid (254.6–337.3 mg/100 g) and stearic
acid (68.3–90.2 mg/100 g) showed a high proportion, and both of these
components increased equally in FeIESLs. The total fatty acid contents
increased after a slight decrease to 601.3, 587.4, and 741.5 mg/100 g of
DrIESLs, StIESLs, and FeIESLs, respectively. The analysis of PCA and
heatmap in Fig. 1 demonstrate specic differences between each sample
at the processing stages. The variability of each of these samples was
signicantly different using the 2D score, with 97.99 % (PC1, 78.35 %;
PC2, 19.64 %) (Fig. 1A) and 95.78 % (PC1, 70.49 %; PC2, 25.29 %)
(Fig. 2B) for saturated and unsaturated fatty acids, respectively. In the
heatmap data shown in Fig. 1C, there were respectively seven compo-
nents that increased or decreased in FeIESLs compared to that in
DrIESLs.
Through fermentation using LAB, lipids can be decomposed,
although not all strains exhibit this capability. Among them, L. brevis,
used in the present study, is known as a hetero fermentative LAB strain
with a lipid-degrading activity (Ziarno et al., 2020). The major fatty
acids identied in DrIESLs were palmitic acid, stearic acid, oleic acid,
and linoleic acid. The fatty acid prole of medium and small soybean
powder also indicated a high proportion of palmitic acid, linoleic acid,
and oleic acid (Lee et al., 2022). In addition, it was conrmed that the
ratio of palmitic acid, stearic acid, and oleic acid was increased in the
fermentation process of white bean soymilk compared to raw bean seeds
(Ziarno et al., 2020). Particularly, linoleic acid decreased during steril-
ization, but increased during fermentation, possibly due to decomposi-
tion under heat stress (Wang et al., 2021). In comparison, arachidic acid,
behenic acid, γ-linolenic acid, eicosenic acid, docosadienoic acid, and
nervonic acid were not detected post-fermentation. This nding may be
attributed to their bioconversion into hydroxy fatty acids by Lactoba-
cillus spp. generated hydratase or decomposition into hydrocarbons,
alcohols, aldehydes, ketones, and aromatic compounds through enzy-
matic action and oxidation (Wang et al., 2021). While saturated fatty
acids are associated with the risks of coronary heart disease, cardio-
vascular diseases, and type 2 diabetes, stearic acid is inversely related
and can increase the content of fatty acid β-oxidation by promoting
mitochondrial fusion and reducing long-chain acylcarnitine (Senyilmaz-
Tiebe et al., 2018). Oleic acid, which exhibited more than a 2.1-fold
increase, exerted benecial effects such as reducing low-density lipo-
protein (LDL) cholesterol, lowering the risk of cardiovascular diseases,
improving glucose regulation and insulin sensitivity, and decreasing
lymphocyte proliferation (Calder, 2015; De Carvalho & Caramujo,
2018). In addition, essential fatty acids such as linoleic acid can lower
serum cholesterol and LDL cholesterol levels (Calder, 2015). Our out-
comes conrmed that fermentation processing of IESLs can obtain the
positive effect of such as oleic acid.
3.3. Comparison of free amino acid contents in the DrIESLs, StIESLs, and
FeIESLs
The results of free amino acid analysis are shown in Table 2 and
Fig. 2. In the free amino acid, except for the non-essential amino acids
citrulline, cystine, β-alanine, and GABA, all other components decreased
in FeIESLs compared to StIESLs. Among them, citrulline signicantly
increased by approximately 6-fold from 7.16 (StIESLs) to 59.50
(FeIESLs) mg/100 g, and β-alanine increased from 47.86 (StIESLs) to
70.39 (FeIESLs) mg/100 g. Furthermore, GABA increased from 308.88
(StIESLs) to 435.48 (FeIESLs) mg/100 g, which accounted for a large
proportion. Although most free amino acids decreased during fermen-
tation, the non-essential free amino acids that accounted for a large
proportion of all amino acids included aspartic acid (from 1169.37 to
993.13 mg/100 g), glutamic acid (from 286.74 to 156.93 mg/100 g),
alanine (from 163.56 to 135.02 mg/100 g), tyrosine (from 105.39 to
75.74 mg/100 g), and arginine (from 246.64 to 177.00 mg/100 g), and
the essential free amino acids included valine (from 219.96 to 181.00
Table 1
Comparison of fatty acid contents in the DrIESLs, StIESLs, and FeIESLs.
Contents
1
(mg/100 g) Processing stages
2
DrIESLs StIESLs FeIESLs
Saturated fatty acids (SFA)
Palmitic acid (C16:0) 254.6 ±
12.3b
257.8 ±
8.3b
337.3 ±
16.9a
Stearic acid (C18:0) 68.3 ±1.4b 65.3 ±1.1c 90.2 ±0.9a
Arachidic acid (C20:0) 5.8 ±0.1b 6.5 ±0.1a nd
3
Behenic acid (C22:0) 11.1 ±0.6a 9.8 ±0.2b nd
Lignoceric acid (C24:0) 14.2 ±0.9a 11.8 ±1.0b 12.4 ±1.0ab
Total 354.0 351.2 439.9
Unsaturated fatty acids (USFA)
Palmitoleic acid (C16:1) 30.9 ±0.9b 36.0 ±1.6a 36.5 ±1.6a
Oleic acid (C18:1n9c) 53.4 ±2.3c 66.3 ±2.2b 113.1 ±7.9a
Linoleic acid (C18:2n6c) 144.5 ±4.6a 112.9 ±
4.2b 136.9 ±3.4a
γ-linolenic acid (C18:3n6) nd 2.9 ±0.1a nd
α
-linolenic acid (C18:3n3) 4.2 ±0.3b 3.7 ±0.3b 8.8 ±0.5a
Eicosenoic acid (C20:1) 3.0 ±0.1a 3.0 ±0.2a nd
Eicosatrienoic acid
(C20:3n3) 4.5 ±0.2b 4.7 ±0.3b 6.3 ±0.4a
Docosadienoic acid (C22:2) 3.3 ±0.1a 3.1 ±0.1a nd
Nervonic acid (C24:1) 3.5 ±0.1a 3.6 ±0.2a nd
Total 247.3 236.2 301.6
Total fatty acids 601.3 587.4 741.5
1
All values are presented as the mean ±SD of pentaplicate determination.
Different letters correspond to the signicant differences relating to samples
using Duncan’s multiple range tests (p <0.05).
2
Processing stages: DrIESLs, fresh isoavone-enriched soybean leaves;
StIESLs, steamed isoavone-enriched soybean leaves; and FeIESLs, fermented
isoavone-enriched soybean leaves.
3
nd: not detected.
H.Y. Lee et al.
Food Chemistry: X 24 (2024) 101999
4
mg/100 g), isoleucine (from 135.41 to 97.75 mg/100 g), leucine (from
133.71 to 91.13 mg/100 g), and phenylalanine (from 178.52 to 124.34
mg/100 g). The total free amino acid contents decreased in the following
order: DrIESLs (3938.98 mg/100 g), StIESLs (3704.55 mg/100 g), and
FeIESLs (3053.41 mg/100 g) (Table 2). The results of the PCA for non-
essential and essential amino acids demonstrate a considerable degree
of variability among the samples during the processing stages. The PC
variability, as indicated by the 2D score plot, revealed that the PC
variability for non-essential amino acids was 92.78 % (PC1, 74.42 %;
PC2, 18.36 %) (Fig. 2A) and that for essential amino acids was 99.10 %
(PC1, 92.41 %; PC2, 6.69 %) (Fig. 2B). In the free amino acid heatmap,
except for the non-essential amino acids citrulline, cystine, β-alanine,
and GABA, all other components decreased in FeIESLs (Fig. 2C).
In our study, most free amino acids tended to decrease after
fermentation, which may be attributed to the proteolytic system for the
growth of LAB and the action of various enzymes such as protease,
peptidase, and amino acid decarboxylase in response to pH stress
(Saubade et al., 2017; Wang et al., 2021; Zhang, Xia, et al., 2023).
Furthermore, LAB exhibited varying auxotrophism across strains, with
some strains necessitating amino acids and peptides as nitrogen sources,
which may elucidate the reduction in free amino acids post-
fermentation (Kieliszek et al., 2021). Even in the dried longan (Dimo-
carpus longan) fermentation, the total free amino acid contents
decreased with a tendency to decrease in most free amino acids, and it
was conrmed that the free amino acid contents decreased in the process
of sterilizing vegetable baby foods (Khan et al., 2018; Kieliszek et al.,
2021). This trend was also observed in mass processing stages in this
study. Since amino acids can also be converted into various avor
compounds, such as aldehydes, alcohols, and esters, analyzing these
compounds may provide insights into the metabolic pathways of amino
acids (Wang et al., 2021). A notable nding in our study was increased
citrulline, β-alanine, and GABA during fermentation. Citrulline is a non-
essential amino acid that serves as the end product of glutamine meta-
bolism as well as a metabolite of arginine. The inverse proportion be-
tween arginine and citrulline observed in DrIESLs and FeIESLs suggested
a potential metabolic interplay. Citrulline offers various health benets,
including enhanced exercise performance, reduced muscle fatigue,
blood pressure regulation, and cholesterol reduction (Papadia et al.,
2018). Carnosine is a combination of β-alanine and histidine, and
anserine is another carnosine derivative, which is a combination of
β-alanine and 1-methyl-histidine. The increase in β-alanine through
lactic acid fermentation may be attributed to uracil or anserine and
carnosine degradation due to the action of enzymes produced during the
fermentation (Hoffman et al., 2015). GABA, a non-protein amino acid,
acts as an inhibitory neurotransmitter in mammals and is synthesized
through the irreversible decarboxylation of L-glutamic acid by gluta-
mate decarboxylase (GAD), with the amino group positioned on the
gamma carbon, comprising four carbons (Lee et al., 2022; Wang et al.,
2021). LAB served as a signicant producer of GABA in IESLs containing
glutamic acid; it could be facilitated by the presence of pyridoxal-5
′
-
phosphate as a cofactor (Lee et al., 2022; Wang et al., 2021). Although
the characteristics of GAD vary across different strains and lineages of
LAB, which mostly indicates the optimal activity of Lactobacillus GAD, is
pH 4.0–5.0, and it is known that the C-terminal region is also involved in
the pH dependence of the catalysis (Pannerchelvan et al., 2023).
Particularly, L. brevis exhibited a higher yield of GABA when compared
to other LABs (Wang et al., 2021). In this study, L. brevis BMK184 may
also have played an important role in producing GABA than L. plantarum
P1201. FeIESLs with GABA-enriched are expected to provide various
health benets, including hypertension prevention, sedative and
Fig. 1. Comparison of fatty acids in the DrIESLs, StIESLs, and FeIESLs. (A) and (B) 2D score plot with principal component analysis, and (C) heatmap analysis.
Processing stages: DrIESLs, dried isoavone-enriched soybean leaves; StIESLs, steamed isoavone-enriched soybean leaves; FeIESLs, fermented isoavone-enriched
soybean leaves.
H.Y. Lee et al.
Food Chemistry: X 24 (2024) 101999
5
diuretic effects, anxiety, and depression relief (Pannerchelvan et al.,
2023).
3.4. Comparison of water-soluble vitamin contents in the DrIESLs,
StIESLs, and FeIESLs
The results of water-soluble vitamin analysis are shown in Fig. 3.
Comparing DrIESLs and FeIESLs for each ingredient, B5 (175.9–131.2
mg/100 g) decreased by approximately 25 % during fermentation, while
C (12.0–0.6 mg/100 g) decreased by 95 %. In contrast, B3 (19.00–130.6
mg/100 g) increased by approximately 6.9-fold, representing a signi-
cant proportion, and B9 (9.7–25.5 mg/100 g) increased by around 2.5-
fold. The total water-soluble vitamin contents were increased after
slightly decreasing to 216.7, 214.3, and 288.0 mg/100 g in the DrIESLs,
StIESLs, and FeIESLs, respectively (Fig. 3A). Fig. 3B provides clear evi-
dence of a stepwise distinction during the processing stages, as indicated
by the PCA. Furthermore, the variability of each axis was 71.78 % for
PC1 and 20.06 % for PC2, demonstrating a signicant change. In the
heatmap depicting changes in the content of water-soluble vitamins,
FeIESLs exhibited a notable pattern of decrease in B5 and C, except for
B2, while B3 and B9 displayed signicant increases (Fig. 3C).
Vitamins are organic compounds that are not completely or sub-
stantially synthesized in the human body, and they are essential
micronutrients for humans as they maintain major cell biochemical re-
actions (LeBlanc et al., 2015; Walther & Schmid, 2017). In our study, we
conrmed the tendency of water-soluble vitamins to increase and
decrease through SLAF of IESLs. Vitamins are known to be produced
through fermentation, and the vitamins generated by various factors
such as fermentation strains, substrates, and conditions are diverse
(LeBlanc et al., 2015; LeBlanc et al., 2020; Walther & Schmid, 2017). In
other words, when a precursor of vitamin is present in the fermentation
substrate, it can be biosynthesized through LAB (Wang et al., 2021). The
precursor of niacin is tryptophan, and, when looking at the change in the
content of free amino acids, it was mostly decreased after fermentation.
It has been reported that niacin is biosynthesized through tryptophan
present in IESLs. Some LAB strains possess genes involved in folate
biosynthesis (Walther & Schmid, 2017; Wang et al., 2021). This process
includes the step of converting the precursor guanosine triphosphate to
tetrahydrofolic acid, and this step could have progressed due to an in-
crease in the folate content in our study (Zhao & Shah, 2016). Some
vitamins increase in content due to the inuence of LAB, and vitamins
are sometimes required for the growth of LAB (LeBlanc et al., 2020).
Fig. 2. Comparison of free amino acids in the DrIESLs, StIESLs, and FeIESLs. (A) and (B) 2D score plot with principal component analysis, and (C) heatmap analysis.
Processing stages: DrIESLs, dried isoavone-enriched soybean leaves; StIESLs, steamed isoavone-enriched soybean leaves; FeIESLs, fermented isoavone-enriched
soybean leaves.
H.Y. Lee et al.
Food Chemistry: X 24 (2024) 101999
6
This effect varies by species and lineage; for instance, in our study, the
ascorbic acid content was decreased. In addition, some vitamins may be
reduced due to heat processing steps involved in the processing and
cooking of food (Senyilmaz-Tiebe et al., 2018). In our fermentation
process, the contents of pantothenic acid and ascorbic acid were also
decreased after the sterilization process. The deciency in increased
folate through lactic acid fermentation can lead to potential risks such as
megaloblastic anemia, neural tube defects seen in congenital malfor-
mations, and low birth weight. The positive effects can reduce the risk of
colon cancer in patients with anti-inammatory and inammatory
bowel disease, and folic acid supplementation is especially recom-
mended for pregnant women (LeBlanc et al., 2020; Walther & Schmid,
2017). In addition, niacin is known to exert benets such as cardio-
vascular disease prevention and stroke reduction, albeit it has not been
evaluated to have any signicant effect in recent clinical trials
(D’Andrea et al., 2019). Our results demonstrated that the fermentation
process of IESLs is particularly associated with increasing niacin and
folic acid levels.
3.5. Comparison of isoavone contents in the DrIESLs, StIESLs, and
FeIESLs
The isoavone chromatogram analysis revealed signicant differ-
ences among the samples, as depicted in Fig. 4. The glycoside iso-
avones increased and then decreased, and malonylglycoside
isoavones continuously decreased. In contrast, the aglycone iso-
avones consistently increased. In Fig. 5A, the contents of glycoside,
malonylglycoside, and aglycone forms for each sample are as follows:
DrIESLs (3053.84, 5478.05, and 391.20
μ
g/g, respectively), StIESLs
(5027.14, 2678.79, and 2271.41
μ
g/g, respectively), and FeIESLs
(735.03, 561.47, and 5121.12
μ
g/g, respectively). The sterilization
process led to a decrease in the malonylglycosides and an increase in the
glycosides and aglycones. Conversely, during fermentation, the content
of malonyl and glycosides decreased, while that of aglycones increased.
The total isoavone contents increased slightly to 8923.09, 9977.34,
and 6417.62
μ
g/g in DrIESLs, StIESLs, and FeIESLs, respectively, and
then decreased. This difference was clearly evident in the PCA and
heatmap. The PCA demonstrated that PC1 and PC2 explained 79.00 %
and 20.81 % of the variation in the data, respectively, thereby high-
lighting the signicant differences between the samples (Fig. 5B).
Particularly, the heatmap results highlighted changes in isoavone de-
rivatives at each processing stage (Fig. 5C). The changes in the
bioconversion mechanism of isoavone compounds according to the
overall process are depicted in Fig. 6.
Isoavones are the avonoid compounds mainly contained in soy-
bean crops and are also called phytoestrogen owing to their chemical
structure, which is similar to that of estrogen (Lee et al., 2018). Gener-
ally, sugars form glucoside bonds and exist in the form of glycosides, and
when ingested, they are decomposed into aglycones due to the enzy-
matic action of intestinal microorganisms, and are only partially
absorbed (Zhang, Zhang, et al., 2023). In other words, it is important not
only to produce DrIESLs by treating ethylene and ethephon, as shown
previously, but also to efciently utilize them, and it is important to
increase their bioavailability through bioconversion to the non-
glycoside form of aglycones (Lee et al., 2023; Yuk et al., 2016). We
found the answer in the SLAF process at a small scale from several
previous research data (Lee et al., 2018; Lee et al., 2022; Zhang, Zhang,
et al., 2023). In our study, during the sterilization step of fermentation,
the isoavones content in malonylglycosides decreased, and that in
glucosides and aglycones increased. Thus, isoavones in the form of
malonyl glucosides are relatively heat-labile, deglycosylated, and
deesteried during the sterilization process (Andrade et al., 2016;
Kuligowski et al., 2022). The increase in total isoavone contents may
be seen as a de-esterication of acetyl glycoside isoavones as well as
malonylglycosides (Lee et al., 2018). In FeIESLs, both malonyl and
glycoside bound forms of isoavones decreased signicantly, while the
aglycone form of isoavones increased markedly. These results
conrmed that the conversion of isoavones was superior in the acid
and β-glucosidase enzymes produced by LAB, even during mass
fermentation processing (Cho et al., 2011;Lee et al., 2018; Lee et al.,
2022). Also, mass fermentation processing suggests that the isoavone
compounds in IESLs may have transformed into other phenolic com-
pounds, such as avonols and phenolic acids (Cho et al., 2011). Mass
processing also, it also improves the overall antioxidant capacity.
Furthermore, it is expected that may have various health benets such as
anti-inammatory, anti-cancer, anti-obesity, anti-osteoporosis, and
menopausal symptom relief (Andrade et al., 2016; Hirose et al., 2016;
Lee et al., 2017; Xie et al., 2021; Zhang, Zhang, et al., 2023).
Table 2
Comparison of free amino acid contents in the DrIESLs, StIESLs, and FeIESLs.
Contents
1
(mg/100 g) Processing stages
2
DrIESLs StIESLs FeIESLs
Non-essential amino acids
Taurine 37.27 ±0.88a 36.32 ±1.44a 32.78 ±1.06b
Aspartic acid 1107.24 ±
40.04a
1169.37 ±
65.81a
993.13 ±
37.85b
Serine 160.40 ±5.73a 138.71 ±2.25b 45.44 ±2.12c
Glutamic acid 251.77 ±9.06b 286.74 ±
20.89a 156.93 ±3.10c
Aminoadipic acid 64.99 ±0.97a 59.20 ±0.97b 56.41 ±1.74c
Glycine 45.30 ±1.46a 42.74 ±0.90b 44.78 ±1.03ab
Alanine 184.82 ±4.85a 163.56 ±3.97b 135.02 ±3.25c
Citrulline 9.87 ±0.17b 7.16 ±0.13c 59.50 ±0.90a
α
-aminobutyric acid 17.29 ±0.57a 15.31 ±0.31b 14.12 ±0.72c
Cystine 10.34 ±0.66b nd
3
15.06 ±0.63a
Cystathionine 15.43 ±0.44a 13.12 ±0.48b 12.01 ±0.48c
Tyrosine 129.47 ±4.28a 105.39 ±3.45b 75.74 ±2.09c
β-alanine 54.62 ±2.69b 47.86 ±1.23c 70.39 ±0.82a
β-aminoisobutyric
acid 84.57 ±1.04a 65.26 ±2.26b 31.96 ±1.33c
γ-aminobutyric acid 357.36 ±6.01b 308.88 ±4.56c 435.48 ±
4.12a
Aminoethanol 38.77 ±0.91a 33.49 ±0.81b 34.74 ±0.35b
Hydroxyproline 13.71 ±0.54a 13.96 ±0.39a 2.68 ±0.26b
Ornithine 3.83 ±0.09a 3.68 ±0.12a 0.42 ±0.01b
1-Methylhistidine 3.53 ±0.13b 5.08 ±0.04a 1.50 ±0.05c
Anserine 27.86 ±0.14a 25.54 ±0.45b 14.92 ±0.28c
Carnosine 8.07 ±0.09b 8.98 ±0.15a 1.59 ±0.05c
Arginine 254.60 ±6.24a 246.64 ±4.66a 177.00 ±
2.70b
Total 2881.11 2796.98 2411.59
Essential amino acids
Threonine 90.41 ±2.22a 76.86 ±0.70b 48.62 ±0.50c
Valine 267.87 ±2.96a 219.96 ±4.68b 181.00 ±2.47c
Methionine 18.34 ±0.40b 22.37 ±0.52a 7.77 ±0.40c
Isoleucine 171.50 ±7.17a 135.41 ±4.34b 97.75 ±2.26c
Leucine 157.86 ±2.61a 133.71 ±3.90b 91.13 ±2.01c
Phenylalanine 199.59 ±2.73a 178.52 ±2.35b 124.34 ±3.17c
Lysine 80.58 ±3.54a 79.50 ±3.11a 41.88 ±0.68b
Histidine 71.71 ±1.72a 61.23 ±0.94b 49.34 ±0.44c
Total 1057.87 907.57 641.82
Total amino acids 3938.98 3704.55 3053.41
Ammonia 113.08 ±4.72b 118.50 ±
3.85ab
124.10 ±
3.75a
Urea 867.15 ±
52.44b
1012.49 ±
78.07a
712.49 ±
53.43c
1
All values are presented as the mean ±SD of pentaplicate determination.
Different letters correspond to the signicant differences relating to samples
using Duncan’s multiple range tests (p <0.05).
2
Processing stages: DrIESLs, fresh isoavone-enriched soybean leaves;
StIESLs, steamed isoavone-enriched soybean leaves; and FeIESLs, fermented
isoavone-enriched soybean leaves.
3
nd: not detected.
H.Y. Lee et al.
Food Chemistry: X 24 (2024) 101999
7
3.6. Comparison of TP and TF contents in the DrIESLs, StIESLs, and
FeIESLs
The changes in the TP and TF contents are depicted in Fig. 7. For
DrIESLs, StIESLs, and FeIESLs, the TP contents were 8.38, 8.44, and
10.37 GAE mg/g, respectively, while the TF contents were 14.37, 15.90,
and 18.35 RE mg/g, respectively (Fig. 7A). These contents commonly
increased slightly after sterilization and increased further during
fermentation. Regarding the alterations in TP and TF, as indicated by
PCA, the difference was minimal for DrIESLs and StIESLs. However,
specic differences were observed for FeIESLs (Fig. 7B).
The phenolic compounds are secondary metabolites commonly pre-
sent in plants and are produced by various factors occurring during plant
growth (Cho et al., 2011). Fermentation is known to increase these
phenolic compounds (Lee et al., 2018; Lee et al., 2022). Particularly,
through the action of hydrolases such as β-glycosidase and decarbox-
ylase, which are produced by bacteria in fermentation using LAB, the
binding phenolic compounds bound to the plant cell wall can be
decomposed, and the glycosylation avonoid compounds can be
decomposed into aglycone forms to form compounds with increased
bioavailability (Cho et al., 2011; Khan et al., 2018; Zhang, Zhang, et al.,
2023). Lactic acid fermentation of dried longan, soybean milk, juju-
be–wolfberry, and African nightshade displayed the same trend of
increasing total phenolic and avonoid contents as in our study (Cho
et al., 2011; Khan et al., 2018; Septembre-Malaterre et al., 2018). It was
also converted into other phenolic compounds via thermal decomposi-
tion during heat processing for sterilization during the fermentation
process, as in the morphological change of isoavones. In our study, the
TP and TF contents changed slightly during the sterilization process and
were signicantly increased during fermentation, suggesting that the
enzyme and metabolism of the used LAB are suitable strains for pro-
ducing and degrading phenolic compounds to produce compounds with
increased bioavailability on mass fermentation processing.
3.7. Comparison of biological activities in the DrIESLs, StIESLs, and
FeIESLs
The results of antioxidant activities (such as DPPH, ABTS, hydroxyl
radical scavenging activities, and FRAP) and digestive enzyme inhibi-
tory activities (including
α
-glucosidase and pancreatic lipase inhibitory
activities) are depicted in Fig. 8. The results measured at a concentration
of 1 mg/mL exhibited that, for DrIESLs, StIESLs, and FeIESLs, respec-
tively, the DPPH radical scavenging activity values were 64.08, 64.58,
and 82.66 % (Fig. 8A), the ABTS radical scavenging activity values were
68.74, 73.52, and 90.21 % (Fig. 8B), the hydroxyl radical scavenging
activity values were 40.12, 44.42, and 54.79 % (Fig. 8C), and the FRAP
values were 0.53, 0.54, and 0.68 OD
525 nm
(Fig. 8D). In addition, similar
antioxidant activities patterns were demonstrated at concentrations of
0.5 and 0.25 mg/mL. Similar to the TP and TF contents, a slight increase
upon sterilization and a more signicant increase upon fermentation
were noted. At a concentration of 1 mg/mL, the
α
-glucosidase inhibitory
activity for DrIESLs, StIESLs, and FeIESLs was 35.13, 38.44, and 50.46
%, respectively (Fig. 8E), while that for pancreatic lipase inhibitory
activity for DrIESLs, StIESLs, and FeIESLs was 43.25, 44.50, and 64.56
Fig. 3. Comparison of water-soluble vitamin contents in DrIESLs, StIESLs, and FeIESLs. (A) Water-soluble vitamin contents, (B) 2D scores plot with principal
component analysis, and (C) heatmap analysis. Processing stages: DrIESLs, dried isoavone-enriched soybean leaves; StIESLs, steamed isoavone-enriched soybean
leaves; FeIESLs, fermented isoavone-enriched soybean leaves. All values are presented as the mean ±SD of pentaplicate determination, and different small letters
correspond to the signicant differences relating to the fermentation time and starter using Duncan’s multiple tests (p <0.05).
H.Y. Lee et al.
Food Chemistry: X 24 (2024) 101999
8
%, respectively (Fig. 8F). In addition, similar digestive enzyme inhibi-
tory activities patterns were noted at concentrations of 0.5 and 0.25 mg/
mL. Both the digestive enzyme inhibitory activities followed the pattern
of slightly increasing after sterilization and then further increasing upon
fermentation, similar to that for the TP, TF contents, and the antioxidant
activities.
We conrmed that the phenolic and avonoid contents increased
through fermentation, indicating a positive correlation with antioxidant
activities that are related to the structure of the phenolic compounds.
The structural characteristic of phenolic compounds is that they have
hydroxyl groups, which prevent free radicals from attacking other cells
by reducing the numbers of hydrogen and electrons to free radicals and
creating a stable state (Sun et al., 2022; Zeb, 2020). Particularly, the
deglycosylated isoavone by β-glucosidase may have increased active
hydroxyl groups, which increases the antioxidant potential (Lee et al.,
2022). As with previous studies, LAB contributed to increasing the
antioxidant activity of IESLs during mass fermentation (Lee et al., 2018;
Papadia et al., 2018; Sun et al., 2022; Zhao & Shah, 2016). The increase
in phenolic compound content and antioxidant activity of soybean
leaves through lactic acid fermentation in a mass process suggests their
commercial potential (Wang et al., 2021; Zeb, 2020). When we consume
foods, some enzymes are activated and involved in the digestion in the
human body. Among them,
α
-glucosidase is an enzyme that acts in the
nal stage of carbohydrate digestion in the mucosa of the small intestine
by hydrolyzing carbohydrates into monosaccharides (Lee et al., 2022;
Nurhayati et al., 2017). Pancreatic lipase is a fat-digesting enzyme that
aids in the absorption of dietary triglycerides (Lee et al., 2022). The
inhibition of such digestive enzymes plays an important role in type 2
diabetes management and anti-obesity activities (Lee et al., 2018; Nur-
hayati et al., 2017). In our study, an increase in digestive enzyme
inhibitory activities was conrmed through the mass fermentation
processing of IESL. The structure of phenolic compounds varies ac-
cording to their type, and the inhibition of digestive enzymes may occur
through irreversible or reversible inhibition (competitive and non-
competitive inhibition) among the compounds of various structures
(Tan et al., 2017). In lactic acid fermentation on mass, increasing the
content of phenolic compounds caused an increase in the digestive
enzyme inhibitory activities (Lee et al., 2018; Sun et al., 2022). There-
fore, mass fermentation processing IESLs can be a good alternative to
drug therapy and synthetic compounds that induce side effects such as
gastrointestinal infection and loss of appetite (Nurhayati et al., 2017).
4. Conclusions
In our study, we observed changes in the nutritional compositions
(including fatty acid, free amino acid, and water-soluble vitamin) and
isoavones of IESLs through mass SLAF using L. plantarum P1201 and
L. brevis BMK184. The content of bioactive compounds, such as oleic
acid, γ-aminobutyric acid, niacin, folic acid, daidzein, and genistein,
increased dramatically after the SLAF of DrIESLs. Particularly, it can be
Fig. 4. Typical isoavones HPLC chromatogram of the 50 % methanol extracts in the DrIESLs, StIESLs, and FeIESLs. (A) Chemical structure of isoavones, (B)
standards, (C) 50 % methanol extracts of DrIESLs, (D) 50 % methanol extracts of StIESLs, and (E) 50 % methanol extracts of FeIESLs. Processing stages: DrIESLs, dried
isoavone-enriched soybean leaves; StIESLs, steamed isoavone-enriched soybean leaves; FeIESLs, fermented isoavone-enriched soybean leaves. 1, daidzin; 2,
genistin; 3, malonyldiadzin; 4, malonylgenistin; 5, daidzein; and 6, genistein.
H.Y. Lee et al.
Food Chemistry: X 24 (2024) 101999
9
Fig. 5. Comparison of isoavones in the DrIESLs, StIESLs, and FeIESLs. (A) Isoavone contents, (B) 2D scores plot with principal component analysis, and (C)
heatmap analysis. Processing stages: DrIESLs, dried isoavone-enriched soybean leaves; StIESLs, steamed isoavone-enriched soybean leaves; FeIESLs, fermented
isoavone-enriched soybean leaves. All values are presented as the mean ±SD of pentaplicate determination, and different small letters correspond to the signicant
differences relating to the fermentation time and starter using Duncan’s multiple tests (p <0.05).
Fig. 6. Bioconversion mechanism of isoavone compounds by the steaming and fermenting process of IESLs. Processing stages: DrIESLs, dried isoavone-enriched
soybean leaves; StIESLs, steamed isoavone-enriched soybean leaves; FeIESLs, fermented isoavone-enriched soybean leaves.
H.Y. Lee et al.
Food Chemistry: X 24 (2024) 101999
10
assumed that these changes are promoted by the action of enzymes such
as GAD and β-glycosidase produced by L. plantarum P1201 and L. brevis
BMK184. Consequently, positive aspects such as the bioconversion of
glycoside isoavones to aglycone isoavones and glutamic acid to
GABA, and increased antioxidant and digestive enzyme inhibitory
activities due to increased phenolic and avonoid contents were
observed. The SLAF system facilitates the effective utilization of IESLs
through articial processing methods. Therefore, by setting the desired
target substance, understanding and appropriately applying the prop-
erties of the fermentation substrate, assessing the impact of processing
Fig. 7. Comparison of total phenolic and total avonoid contents in the DrIESLs, StIESLs, and FeIESLs. (A) Total phenolic and avonoid contents, and (B) 2D scores
plot with principal component analysis. Processing stages: DrIESLs, dried isoavone-enriched soybean leaves; StIESLs, steamed isoavone-enriched soybean leaves;
FeIESLs, fermented isoavone-enriched soybean leaves. All values are presented as the mean ±SD of pentaplicate determination, and different small letters
correspond to the signicant differences relating to the fermentation time and starter using Duncan’s multiple tests (p <0.05).
Fig. 8. Comparison of antioxidant and digestive enzyme inhibitory activities in the DrIESLs, StIESLs, and FeIESLs. (A) DPPH radical-scavenging activity, (B) ABTS
radical-scavenging activity, (C) hydroxyl radical-scavenging activity, (D) ferric reducing/antioxidant power, (E)
α
-glucosidase inhibitory activity, and (F) pancreatic
lipase inhibitory activity. Processing stages: DrIESLs, dried isoavone-enriched soybean leaves; StIESLs, steamed isoavone-enriched soybean leaves; FeIESLs, fer-
mented isoavone-enriched soybean leaves. All values are presented as the mean ±SD of pentaplicate determination, and different small letters correspond to the
signicant differences relating to the fermentation time and starter using Duncan’s multiple tests (p <0.05).
H.Y. Lee et al.
Food Chemistry: X 24 (2024) 101999
11
methods and environmental factors, and analyzing the positive aspects
that can be obtained therefrom, materials with enhanced nutritional
value can be secured and utilized efciently. Finally, L. plantarum P1201
and L. brevis BMK184, along with mass IESL, demonstrated positive
synergistic effects, which suggest their potential utility as functional
food material in commercial market.
CRediT authorship contribution statement
Hee Yul Lee: Writing – review & editing, Writing – original draft,
Visualization, Project administration. Ji Ho Lee: Writing – original
draft, Methodology, Investigation, Data curation. Du Yong Cho: Vali-
dation, Software, Resources. Kyeong Jin Jang: Investigation, Formal
analysis, Data curation. Jong Bin Jeong: Investigation, Formal analysis.
Min Ju Kim: Formal analysis, Data curation. Ga Young Lee: Formal
analysis, Data curation. Mu Yeun Jang: Formal analysis, Data curation.
Jin Hwan Lee: Writing – review & editing, Funding acquisition,
Conceptualization. Kye Man Cho: Writing – review & editing, Super-
vision, Resources, Project administration, Funding acquisition,
Conceptualization.
Declaration of competing interest
The authors declare that they have no known competing nancial
interests or personal relationships that could have appeared to inuence
the work reported in this paper.
Acknowledgments
This research was supported by the Basic Science Research Program
through the National Research Foundation (NRF) funded by the Ministry
of Education (Grant number 2016R1D1A1B01009898, RS-2023-
00245096, and RS-2024-00346785), Republic of Korea.
Appendix A. Abbreviation used
The Abbreviation used in the study are DrIESLs, dried isoavone-
enriched soybean leaves; StIESLs, steamed isoavone-enriched soy-
bean leaves; and FeIESLs, fermented isoavone-enriched soybean
leaves.
Appendix B. Supplementary data
Supplementary data to this article can be found online at https://doi.
org/10.1016/j.fochx.2024.101999.
Data availability
Data will be made available on request.
References
Andrade, J. C., Mandarino, J. M. G., Kurozawa, L. E., & Ida, E. I. (2016). The effect of
thermal treatment of whole soybean our on the conversion of isoavones and
inactivation of trypsin inhibitors. Food Chemistry, 194, 1095–1101. https://doi.org/
10.1016/j.foodchem.2015.08.115
Basu, P., & Maier, C. (2018). Phytoestrogens and breast cancer: In vitro anticancer
activities of isoavones, lignans, coumestans, stilbenes and their analogs and
derivatives. Biomedicine & Pharmacotherapy, 107, 1648–1666. https://doi.org/
10.1016/j.biopha.2018.08.100
Calder, P. C. (2015). Functional roles of fatty acids and their effects on human health.
Journal of Parenteral and Enteral Nutrition, 39, 18S–32S. https://doi.org/10.1177/
0148607115595980
Campos, M. G. (2021). Soy Isoavones. In J. Xiao, S. D. Sarker, & Y. Asakawa (Eds.),
Handbook of dietary phytochemicals (pp. 205–242). Singapore: Springer. https://doi.
org/10.1007/978-981-15-4148-3_8.
Chen, L., Teng, H., Xie, Z., Cao, H., Cheang, W. S., Skalicka-Woniak, K., & Xiao, J. (2018).
Modications of dietary avonoids towards improved bioactivity: An update on
structure–activity relationship. Critical Reviews in Food Science and Nutrition, 58(4),
513–527. https://doi.org/10.1080/10408398.2016.1196334
Cho, K. M., Lee, J. H., Yun, H. D., Ahn, B. Y., Kim, H., & Seo, W. T. (2011). Changes of
phytochemical constituents (isoavones, avanols, and phenolic acids) during
cheonggukjang soybeans fermentation using potential probiotics Bacillus subtilis CS90.
Journal of Food Composition and Analysis, 24(3), 402–410. https://doi.org/10.1016/j.
jfca.2010.12.015
D’Andrea, E., Hey, S. P., Ramirez, C. L., & Kesselheim, A. S. (2019). Assessment of the
role of niacin in managing cardiovascular disease outcomes: A systematic review and
meta-analysis. JAMA Network Open, 2(4), e192224. https://doi.org/10.1001/
jamanetworkopen.2019.2224
De Carvalho, C. C., & Caramujo, M. J. (2018). The various roles of fatty acids. Molecules,
23(10), 2583. https://doi.org/10.3390/molecules23102583
Hirose, A., Terauchi, M., Akiyoshi, M., Owa, Y., Kato, K., & Kubota, T. (2016). Low-dose
isoavone aglycone alleviates psychological symptoms of menopause in Japanese
women: A randomized, double-blind, placebo-controlled study. Archives of
Gynecology and Obstetrics, 293, 609–615. https://doi.org/10.1007/s00404-015-
3849-0
Hoffman, J. R., Landau, G., Stout, J. R., Hoffman, M. W., Shavit, N., Rosen, P., &
Ostfeld, I. (2015). β-Alanine ingestion increases muscle carnosine content and
combat specic performance in soldiers. Amino Acids, 47, 627–636. https://doi.org/
10.1007/s00726-014-1896-7
Khan, S. A., Liu, L., Lai, T., Zhang, R., Wei, Z., Xiao, J., & Zhang, M. (2018). Phenolic
prole, free amino acids composition and antioxidant potential of dried longan
fermented by lactic acid bacteria. Journal of Food Science and Technology, 55(12),
4782–4791. https://doi.org/10.1007/s13197-018-3411-8
Kieliszek, M., Pobiega, K., Piwowarek, K., & Kot, A. M. (2021). Characteristics of the
proteolytic enzymes produced by lactic acid bacteria. Molecules, 26(7), 1858.
https://doi.org/10.3390/molecules26071858
Kim, M. S., Jung, Y. S., Jang, D., Cho, C. H., Lee, S. H., Han, N. S., & Kim, D. O. (2022).
Antioxidant capacity of 12 major soybean isoavones and their bioavailability under
simulated digestion and in human intestinal Caco-2 cells. Food Chemistry, 374,
Article 131493. https://doi.org/10.1016/j.foodchem.2021.131493
Kuligowski, M., Sobkowiak, D., Polanowska, K., & Jasi´
nska-Kuligowska, I. (2022). Effect
of different processing methods on isoavone content in soybeans and soy products.
Journal of Food Composition and Analysis, 110, Article 104535. https://doi.org/
10.1016/j.jfca.2022.104535
Latimer, G. W. (2012). AOAC international. Ofcial methods of analysis of AOAC
international (19th ed.). Gaithersburg, MD, USA: AOAC International.
LeBlanc, J. G., Lai˜
no, J. E., del Valle, M. J., de Giori, G. S., Sesma, F., & Taranto, M. P.
(2015). B-group vitamins production by probiotic lactic acid bacteria. Biotechnology
of lactic acid bacteria: Novel applications, 279–296. https://doi.org/10.1002/
9781118868386.ch17
LeBlanc, J. G., Levit, R., Savoy de Giori, G., & De Moreno De LeBlanc, A. (2020).
Application of vitamin-producing lactic acid bacteria to treat intestinal inammatory
diseases. Applied Microbiology and Biotechnology, 104, 3331–3337. https://doi.org/
10.1007/s00253-020-10487-1
Lee, H., Choue, R., & Lim, H. (2017). Effect of soy isoavones supplement on climacteric
symptoms, bone biomarkers, and quality of life in Korean postmenopausal women: A
randomized clinical trial. Nutrition Research and Practice, 11(3), 223–231. https://
doi.org/10.4162/nrp.2017.11.3.223
Lee, H. Y., Cho, D. Y., Jang, K. J., Lee, J. H., Jung, J. G., Kim, M. J., & Cho, K. M. (2022).
Changes of γ-aminobutyric acid, phytoestrogens, and biofunctional properties of the
isoavone-enriched soybean (Glycine max) leaves during solid lactic acid
fermentation. Fermentation, 8(10), 525. https://doi.org/10.3390/
fermentation8100525.
Lee, J. H., Cho, D. Y., Jang, K. J., Jeong, J. B., Lee, G. Y., Jang, M. Y., & Cho, K. M. (2023).
Changes in nutrient components and digestive enzymatic inhibition activities in soy
leaves by ethephon treatment. Plants, 12(20), 3640. https://doi.org/10.3390/
plants12203640
Lee, J. H., Kim, B., Hwang, C. E., Haque, M. A., Kim, S. C., Lee, C. S., & Lee, D. H. (2018).
Changes in conjugated linoleic acid and isoavone contents from fermented
soymilks using Lactobacillus plantarum P1201 and screening for their digestive
enzyme inhibition and antioxidant properties. Journal of Functional Foods, 43, 17–28.
https://doi.org/10.1016/j.jff.2018.01.022
Liu, H., Wang, Y., Zhu, D., Xu, J., Xu, X., & Liu, J. (2023). Bioaccessibility and application
of soybean isoavones: A review. Food Reviews International, 39(8), 5948–5967.
https://doi.org/10.1080/87559129.2022.2103824
Lu, F., Wang, Y., Wu, S., Huang, W., Yao, H., Wang, S., & Herrera-Balandrano, D. D.
(2024). Germination time and in vitro gastrointestinal digestion impact on the
isoavone bioaccessibility and antioxidant capacities of soybean sprouts. Food
Chemistry, 460, Article 140517. https://doi.org/10.1016/j.foodchem.2024.140517
Nurhayati, R., Frediansyah, A., & Rachmah, D. L. (2017, December). Lactic acid bacteria
producing inhibitor of alpha glucosidase isolated from ganyong (Canna edulis) and
kimpul (Xanthosoma sagittifolium). In IOP conference series: Earth and environmental
science. IOP Publishing. https://doi.org/10.1088/1755-1315/101/1/012009. Vol.
Vol. 101, No. 1, p. 012009.
Okabe, Y., Shimazu, T., & Tanimoto, H. (2011). Higher bioavailability of isoavones
after a single ingestion of aglycone-rich fermented soybeans compared with
glucoside-rich non-fermented soybeans in Japanese postmenopausal women. Journal
of the Science of Food and Agriculture, 91(4), 658–663. https://doi.org/10.1002/
jsfa.4228
Pannerchelvan, S., Rios-Solis, L., Wong, F. W. F., Zaidan, U. H., Wasoh, H.,
Mohamed, M. S., & Halim, M. (2023). Strategies for improvement of gamma-
aminobutyric acid (GABA) biosynthesis via lactic acid bacteria (LAB) fermentation.
Food & Function, 14(9), 3929–3948. https://doi.org/10.1039/D2FO03936B
H.Y. Lee et al.
Food Chemistry: X 24 (2024) 101999
12
Papadia, C., Osowska, S., Cynober, L., & Forbes, A. (2018). Citrulline in health and
disease. Review on human studies. Clinical Nutrition, 37(6), 1823–1828. https://doi.
org/10.1016/j.clnu.2017.10.009
Saubade, F., Hemery, Y. M., Guyot, J. P., & Humblot, C. (2017). Lactic acid fermentation
as a tool for increasing the folate content of foods. Critical Reviews in Food Science and
Nutrition, 57(18), 3894–3910. https://doi.org/10.1080/10408398.2016.1192986
Senyilmaz-Tiebe, D., Pfaff, D. H., Virtue, S., Schwarz, K. V., Fleming, T., Altamura, S., &
Teleman, A. A. (2018). Dietary stearic acid regulates mitochondria in vivo in
humans. Nature Communications, 9(1), 1–10. https://doi.org/10.1038/s41467-018-
05614-6
Septembre-Malaterre, A., Remize, F., & Poucheret, P. (2018). Fruits and vegetables, as a
source of nutritional compounds and phytochemicals: Changes in bioactive
compounds during lactic fermentation. Food Research International, 104, 86–99.
https://doi.org/10.1016/j.foodres.2017.09.031
Sun, X., Zhang, Y., Li, F., Jiao, X., Ma, D., Zhang, L., & Li, Q. (2022). Effects of lactic acid
bacteria fermentation on chemical compounds, antioxidant capacities and
hypoglycemic properties of pumpkin juice. Food Bioscience, 50, Article 102126.
https://doi.org/10.1016/j.fbio.2022.102126
Tan, Y., Chang, S. K., & Zhang, Y. (2017). Comparison of
α
-amylase,
α
-glucosidase and
lipase inhibitory activity of the phenolic substances in two black legumes of different
genera. Food Chemistry, 214, 259–268. https://doi.org/10.1016/j.
foodchem.2016.06.100
Walther, B., & Schmid, A. (2017). Effect of fermentation on vitamin content in food. In
Fermented foods in health and disease prevention (pp. 131–157). Academic Press.
https://doi.org/10.1016/B978-0-12-802309-9.00007-8.
Wang, Y., Wu, J., Lv, M., Shao, Z., Hungwe, M., Wang, J., & Geng, W. (2021). Metabolism
characteristics of lactic acid bacteria and the expanding applications in food
industry. Frontiers in Bioengineering and Biotechnology, 9, Article 612285. https://doi.
org/10.3389/fbioe.2021.612285
Xie, C. L., Park, K. H., Kang, S. S., Cho, K. M., & Lee, D. H. (2021). Isoavone-enriched
soybean leaves attenuate ovariectomy-induced osteoporosis in rats by anti-
inammatory activity. Journal of the Science of Food and Agriculture, 101(4),
1499–1506. https://doi.org/10.1002/jsfa.10763
Yuk, H. J., Song, Y. H., Curtis-Long, M. J., Kim, D. W., Woo, S. G., Lee, Y. B., & Park, K. H.
(2016). Ethylene induced a high accumulation of dietary isoavones and expression
of isoavonoid biosynthetic genes in soybean (Glycine max) leaves. Journal of
Agricultural and Food Chemistry, 64(39), 7315–7324. https://doi.org/10.1021/acs.
jafc.6b02543
Zeb, A. (2020). Concept, mechanism, and applications of phenolic antioxidants in foods.
Journal of Food Biochemistry, 44(9), Article e13394. https://doi.org/10.1111/
jfbc.13394
Zhang, B., Deng, Z., Ramdath, D. D., Tang, Y., Chen, P. X., Liu, R., & Tsao, R. (2015).
Phenolic proles of 20 Canadian lentil cultivars and their contribution to antioxidant
activity and inhibitory effects on
α
-glucosidase and pancreatic lipase. Food Chemistry,
172, 862–872. https://doi.org/10.1016/j.foodchem.2014.09.144
Zhang, J., Xia, N., Teng, J., Wei, B., & Huang, L. (2023). Effect of lactic acid bacteria
fermentation on extractable and non-extractable polyphenols of soybean milk:
Inuence of β-glucosidase and okara. Food Bioscience, 56, Article 103326. https://
doi.org/10.1016/j.fbio.2023.103326
Zhang, K., Zhang, T. T., Guo, R. R., Ye, Q., Zhao, H. L., & Huang, X. H. (2023). The
regulation of key avor of traditional fermented food by microbial metabolism: A
review. Food Chemistry: X. , Article 100871. https://doi.org/10.1016/j.
fochx.2023.100871
Zhao, D., & Shah, N. P. (2016). Lactic acid bacterial fermentation modied phenolic
composition in tea extracts and enhanced their antioxidant activity and cellular
uptake of phenolic compounds following in vitro digestion. Journal of Functional
Foods, 20, 182–194. https://doi.org/10.1016/j.jff.2015.10.033
Ziarno, M., Bry´
s, J., Parzyszek, M., & Veber, A. (2020). Effect of lactic acid bacteria on
the lipid prole of bean-based plant substitute of fermented milk. Microorganisms, 8
(9), 1348. https://doi.org/10.3390/microorganisms8091348
H.Y. Lee et al.
Food Chemistry: X 24 (2024) 101999
13
