Food Chemistry 264 (2018) 189–198

Contents lists available at ScienceDirect

Food Chemistry
journal homepage: www.elsevier.com/locate/foodchem

Fermentation and complex enzyme hydrolysis for improving the total T

soluble phenolic contents, flavonoid aglycones contents and bio-activities of
guava leaves tea
⁎
Lu Wanga, You Luoa, Yanan Wua,b, Yan Liua, Zhenqiang Wua,
a
School of Biology and Biological Engineering, South China University of Technology, Guangzhou 510006, PR China
b
Jiangmen Nanyue Guava Tea Farmer Cooperatives, Jiangmen, Guangdong 529000, PR China

A R T I C LE I N FO A B S T R A C T

Chemical compounds studied in this article: There are both soluble and insoluble-bound forms of phenolics in tea-leaf products. In order to increase total
Gallic acid (Pubchem CID: 370) soluble phenolics contents, guava leaves tea (GLT) was first fermented with Monascus anka and Saccharomyces
Chlorogenic acid (Pubchem CID: 1794427) cerevisiae, and then hydrolyzed with complex enzymes. The changes in phenolics profiles, antioxidant activities
p-Hydroxybenzoic acid (Pubchem CID: 315) and inhibitory effect on α-glucosidase in processed GLT were investigated. Compared with the un-fermented
Ferulic acid (Pubchem CID: 445858)
GLT, fermentation and complex enzymatic processing (FE) significantly increased the total phenolics, total
Coumaric acid (Pubchem CID: 637542)
flavonoids, quercetin and kaempferol contents by 2.1, 2.0, 13.0 and 6.8 times, respectively. After the FE, a major
Caffeic acid (Pubchem CID: 689043)
Syringic acid (Pubchem CID: 10742) proportion of phenolics existed in the soluble form. Quercetin was released in the highest amount among dif-
Rutin (Pubchem CID: 5280805) ferent phenolics. In addition, soluble phenolic extracts from GLT following FE exhibited a highest antioxidant
Isoquercitrin (Pubchem CID: 5280804) activity and inhibitory effect on α-glucosidase. The paper suggested an improved method for processing GLT into
Quercetin-3-O-β-D-xylopyranoside (Pubchem high-value products rich in phenolics and flavonoids aglycones with enhanced health benefits.
CID: 5320861)
Quercetin-3-O-α-L-arabinoside (Pubchem CID:
5481224)
Avicularin (Pubchem CID: 5490064)
Quercetin (Pubchem CID: 5280343)
Quercitrin (Pubchem CID: 5280459)
Kaempferol (Pubchem CID: 5280863)
Keywords:
Guava leaves tea
Phenolics compositions
Fermentation
Enzymatic hydrolysis
Antioxidant activity
α-Glucosidase inhibition activity

1. Introduction phenolics (Díaz-de-Cerio, Gómez-Caravaca, Verardo, Fernández-

Polyphenolic compounds have attracted much attention due to their
various potential health benefits, such as preventing reactive oxygen
species damage, diabetes, and DNA mutations (Albishi, John, Al-
Khalifa, & Shahidi, 2013; Ranilla, Kwon, Apostolidis, & Shetty, 2010).
Guava leaves (Psidium guajava L.) tea (GLT) are an important source of
Gutiérrez, & Segura-Carretero, 2016; Wang, Bei, Wu, Liao, & Wu,
2017). Due to its unique flavor and important nutritional value, there
has been increasing interest in utilizing guava leaf as a functional tea or
raw materials of functional beverages.
Phenolics are existed in soluble and insoluble-bound forms in food
or tea products (de Camargo et al., 2016; Van Hung, 2016). Previous

Abbreviations: GLT, guava leaves tea; SSF, solid state fermentation; DM, dried mass; UF, unfermentation processing; F, fermentation processing; FE, fermentation and enzymatic
hydrolysis processing; SPUF, soluble phenolics of unfermented GLT; SPF, soluble phenolics of fermented GLT; SPFE, soluble phenolics of fermentation and enzymatic hydrolysis GLT;
IBPUF, insoluble-bound phenolics of unfermented GLT; IBPF, insoluble-bound phenolics of fermented GLT; IBPFE, insoluble-bound phenolics of fermentation and enzymatic hydrolysis
GLT; DPPH, 1,1-diphenyl-2-picrylhydrazyl; ABTS, 2,2-azino-bis (3-ethylbenzothiazoline-6-sulfonic acid) diammonium salt; Vc, ascorbic acid
⁎
Corresponding author.
E-mail address: btzhqwu@scut.edu.cn (Z. Wu).

https://doi.org/10.1016/j.foodchem.2018.05.035
Received 7 November 2017; Received in revised form 29 April 2018; Accepted 4 May 2018
Available online 08 May 2018
0308-8146/ © 2018 Elsevier Ltd. All rights reserved.
L. Wang et al.
studies on the phenolic profiles of guava leaf based their analyses on
extracts obtained from different hydrophilic solvents, such as ethyl
acetate, ethanol, methanol, and acetone (Wang, Hwang, Lee, & Lim,
2016). However, these procedures are not sufficient to completely ex-
tract insoluble-bound phenolic compounds, including those conjugated
to small peptides, proteins, polysaccharides or oligosaccharides (Yeo &
Shahidi, 2017). Currently, microbial fermentation is an increasingly
interesting biotechnological processing method that is used to enhance
the nutritional value and organoleptic qualities of tea products by
breaking the bond between phenolics and other substituents in con-
jugated molecules based on the enzymes produced by microorganisms
(Dulf, Vodnar, & Socaciu, 2016; Liu et al., 2017). For example, fer-
mentation with fungi was applied to enhance the antioxidant capacity
of plum by-products, oats and black beans (Dey, Chakraborty, Jain,
Sharma, & Kuhad, 2016; Dulf et al., 2016). Though Aspergillus niger is
used as the starter for solid state fermentation (SSF) and has a great
potential to increase the antioxidant activity of food or tartary buck-
wheat leaves (Dulf, Vodnar, Dulf, & Tosa, 2015; Zhang et al., 2017), it
cannot be widely used to process tea products due to its bad infusion
color. Monascus anka is a widely used food-grade microbial starter that
has a great potential to enhance the bio-activity by releasing insoluble-
bound phenolics (Bei, Liu, Wang, Chen, & Wu, 2017; Wang et al.,
2017). Moreover, it can also produce natural functional bioactive
compounds, such as γ-aminobutyric acid, Monacolin K, and pigments
(Dikshit, & Tallapragada, 2016). It has been reported that quercetin
aglycones possessed a higher inhibition effect towards α-glucosidase or
α-amylase and can decrease postprandial hyperglycemia by retarding
the absorption of glucose (Bhandari, Jong-Anurakkun, Hong, &
Kawabata, 2008). Many researches have also confirmed that the anti-
oxidant activities of flavonoid aglycones (quercetin and kaempferol)
were much higher than those of the glycosides forms (de Araújo et al.,
2013; Manach et al., 1997). It may be due to the complexity of plant cell
structures and the limitation of some specific enzymes (xylanase and β-
glucosidase) produced by M. anka, it was found that fermented guava
leaves still included most of the flavonoid compounds that existed in
the quercetin glycosides forms. Consequently, complex enzymes con-
sisted of cellulase, β-glucosidase, xylanase and hemicellulase were se-
lected to carry out the study for further releasing the phenolics com-
pounds, and converting the flavonoid glycosides into aglycones.
In the present study, GLT were first fermented using food grade
Monascus anka and Saccharomyces cerevisiae, and then hydrolyzed with
complex enzymes. The aim of this study was to investigate the changes
in the soluble and insoluble-bound phenolics from GLT and their anti-
oxidant activity, reducing power and inhibition potential towards α-
glucosidase using different processing methods. Moreover, the changes
in the compositions and contents of individual phenolic compounds in
the soluble and insoluble-bound forms from GLT were also analyzed.
This research aimed to provide guidance for processing tea products
with the stronger bio-activities by improving soluble phenolics and
flavonoid aglycones.
2. Materials and methods
2.1. Chemicals and plant material
Commercial GLT were provided by the Jiangmen Nanyue Guava Tea
farmer cooperatives (Jiangmen, Guangdong, China). L-Ascorbic acid
(Vc), 2-azino-bis (3-ethylbenzothiazoline-6-sulfonic acid) diammonium
salt (ABTS, > 99.7%), N-(1-naphthyl) ethylenediamine dihy-
drochloride, potassium ferricyanide [K3Fe(CN)6], trichloroacetic acid,
K2S2O8 (> 99.8%), Folin-Ciocalteu phenol reagent (> 99.8%), and 1,1-
diphenyl-2-picrylhydrazyl (DPPH, > 99.7%) were all purchased from
Aladdin Reagents Co., Ltd. (Pudong New Area, Shanghai, China).
NaNO2, FeCl3, sodium hydroxide (NaOH), potassium phosphate
monobasic (KH2PO4) and potassium phosphate dibasic (K2HPO4) were
purchased from Tianjin Kemiou Chemical Reagent Co., Ltd. (Jinnan,
190
Food Chemistry 264 (2018) 189–198
Tianjin, China). Individual phenolic acid, flavonoid standards, p-ni-
trophenyl-α-D-glucopyranoside (99.8%), 6-hydroxy-2,5,7,8-tetra-
methylchroman-2-carboxylic acid (Trolox), and α-glucosidase (EC
3.2.1.20) (≥99.8%) were purchased from Sigma-Aldrich (St. Louis,
Missouri, USA). Formic acid and acetonitrile solvents were purchased
from Fisher Scientific Co., Ltd. (HPLC grade, 99.9%, Waltham,
Massachusetts, USA). Complex enzymes including xylanase (E.C.
3.2.1.8, 8000 U/g), cellulase (E.C. 3.2.1.4, 8000 U/g), and hemi-
cellulase (8000 U/g) both from Aspergillus niger and β-glucosidase (E.C.
3.2.1.21, 8000 U/g) from Trichoderma reesei, were food grade and were
purchased from Youtell Biochemical Com., Ltd (Pudong New Area,
Shanghai, China).
2.2. Microorganisms
Monascus anka GIM 3.592 and Saccharomyces cerevisiae GIM 2.139
(deposited in the publicly accessible culture collection GIMCC/GDMCC,
Guangdong Culture Collection Centre of Microbiology, China) were
used in this study. M. anka GIM 3.592 was cultivated on PDA medium
at 30 °C for 7 days before use. The liquid seed medium for M. anka in-
cluded (g/L) peptone 10, glucose 20, KH2PO4 4, yeast extract 3,
FeSO4·7H2O 0.01 and KCl 0.5. Seed cultivation of M. anka was con-
ducted in a 250 mL Erlenmeyer flask with 50 mL of media for 26 h at
30 °C and 180 rpm. The seed medium for S. cerevisiae GIM 2.139
(GIMCC/GDMCC) consisted of (g/L) peptone 20, glucose 20, and NaCl
10 at pH 7.2–7.4. Cultivation of inoculum for S. cerevisiae was prepared
in a 250 mL Erlenmeyer flask with 50 mL of media for 20 h at 30 °C and
180 rpm.
2.3. Processing of guava leaves
The composition of the SSF substrate media was GLT (25%, w/w,
air-dried), rice flour (15%, w/w), the seed cultivation of S. cerevisiae (8
log Colony-Forming Units/mL, 5%, v/w), and the seed cultivation of M.
anka (1 × 106 spores/mL, 10%, v/w) with a water content of 45% (w/
w). Before fermentation, the mixture of GLT substrate was steam-ster-
ilized at 121 °C for 15 min. After thoroughly mixing with the inoculum,
the SSF media was incubated on a round screen (20-mesh, GB/T6003.1,
Guangzhou, China) at 28 °C for 8 days under 45% relative humidity. An
SSF of the substrate without inoculation was applied as a control. All
experiments were performed in triplicate. After SSF, 2% complex en-
zymes (cellulase:xylanase:hemicellulase:β-glucosidase; 1:1:1:1) were
mixed with the above fermented GLT solid state substrates, and 20 mL
of H2O (adjusted to pH = 5.0 using citric acid) was added to the mix-
ture, which was incubated for 12 h at 50 °C. Enzymatic processing was
performed in triplicate. After completion of the reaction, all samples
were maintained at oven for 20 min at 80 °C to inactivate the enzyme.
All samples following the above different processing were collected and
dried for 15 h at 60 °C.
2.4. Extraction of soluble phenolic fractions
The soluble phenolic fractions were extracted according to the re-
ports described by Qiu, Liu, and Beta. (2010). Briefly, 1 g of the dried
GLT samples following the above different processing, which ground to
40-mesh using a micromill, was blended with 80% methanol at a ratio
of 1:10 (w/v). The mixture was kept in a water bath at 40 °C for 1 h. The
filtrate was extracted two times with 80% methanol. The pooled ex-
traction was cleared with a rotary evaporator at 45 °C, and the extract
was re-dissolved in 5 mL of 80% methanol (v/v) to yield the soluble
phenolic fractions.
2.5. Extraction of insoluble bound phenolics fractions
The insoluble bound phenolic fractions were extracted from the
above residue using the basic methods described by Zhang et al. (2010).
L. Wang et al.
After the soluble phenolic extraction, the residue was first dried for 10 h
at 60 °C, and then, the mass was recorded. A total of 0.5 g of the dried
residue was hydrolyzed by adding 50 mL of 2 M NaOH at 25 °C for 4 h.
The mixture was then acidified with concentrated hydrochloric acid to
pH 2.0. The remaining mixture was then extracted three times with
70 mL of ethyl acetate. After the removal of the pooled ethyl acetate,
the bound polyphenolics were dissolved with 5 mL of 80% methanol (v/
v), which was stored at −20 °C before analysis.
2.6. Determinations of total phenolic and total flavonoid contents
The two forms of phenolic fractions (soluble and insoluble-bound)
were determined based on the reported method, with minor modifica-
tion (Singleton, Orthofer, & Lamuela-Raventos, 1999). In brief, the re-
action system consisted of Folin–Ciocalteu reagent (0.3 mL), 20%
Na2CO3 (1.5 mL) and sample extractions or gallic acid standard solution
(1 mL); then, the reaction was incubated at 35 °C for 30 min. Purified
water instead of sample extractions were used as blanks. The absor-
bance was measured using a SpectraMax Gemini microtiter plate reader
(Molecular Devices, Sunnyvale, CA) at a wavelength of 760 nm (A760).
A standard curve was plotted using gallic acid (10, 20, 40, 60, 80, and
100 μg/mL) as a standard (R2 = 0.9995). The contents of soluble and
insoluble bound phenolics were all expressed as mg gallic acid
equivalents (GAE)/g sample in dry mass (DM). The flavonoid content
was determined using a previously described method with slight
modification (Cai et al., 2011). Briefly, 0.1 mL of the above extract was
placed in a 2-mL Eppendorf tube. A 70% methanol solution was added
to make a 0.5 mL solution, and then, 30 μL of 5% NaNO2 solution was
added. After being kept at room temperature for 5 min, 30 μL of the
10% AlCl3 solution was added, and the mixture was incubated for an-
other 6 min. Then, a total of 0.2 mL of 1 M NaOH was added, and the
total volume was increased to 1 mL using a 70% methanol solution. The
solution was thoroughly mixed again and allowed to stand for 30 min at
35 °C. The absorbance was read at 510 nm using the 70% methanol
solution as a blank. The flavonoid contents were expressed as mg rutin
equivalents (RE)/g sample in dry mass. The calibration curves of rutin
were prepared with a standard chemical content between 0.01 and
0.1 mg/mL (R2 = 0.9997). All samples were tested in triplicate. All
reported values are presented as the mean ± standard deviation (SD).
2.7. HPLC analysis of phenolics compounds
A Waters e2695 HPLC system equipped with a Waters Sunfire C18
plus column (250 mm × 4.6 mm, 5 μm, Waters, USA) as well as a Diode
Array Detector (DAD, Waters 2998, Milford, MA, USA) was used to
analyze the compositions and contents of the phenolic compounds in
GLT at different processing stages. The mobile phase included a 0.1%
formic acid aqueous solution (v/v, solution A) and a acetonitrile solu-
tion (solution B) at a constant flow rate of 0.8 mL/min with the fol-
lowing gradient program: 0–5 min, 15% solution B; 5–10 min, 15–20%
solution B; 10–20 min, 20–25% solution B; 20–30 min, 25–35% solution
B; 30–40 min, 35–50% solution B; 40–50 min, 80% solution B; and
50–55 min, 15% solution B. The other chromatographic conditions in-
cluded all wavelength scanning detection (200–600 nm), an injection
volume of 10 μL, and a column temperature of 30 °C. The samples were
filtered through a 0.25-μm membrane filter (Millipore, Billerica, MA,
USA) prior to HPLC analysis. Individual analytes were identified based
on the chromatography of authentic standards and reference data
(Díaz-de-Cerio et al., 2016; Wang et al., 2017). For quantitative ana-
lysis, the recorded wavelength of the chromatographic data was set
under the maximum absorption of the compounds. The concentration of
each analyte was calculated based on a standard curve, and the results
are expressed as milligram per 100 g DM of samples (Table S1). The
precision of the instrument, the repeatability of the method and the
stability of the samples were evaluated based on observed RSD of the
retention times and peak areas of the sixteen analytes from five
191
Food Chemistry 264 (2018) 189–198
repeated runs of GLT samples extracts. The recovery rates were calcu-
lated based on comparisons between results from the spiked and un-
spiked analytes (Table S2).
2.8. Determination of antioxidant activity and reducing power
2.8.1. DPPH radical scavenging activity
Determination of the DPPH radical scavenging activity was per-
formed according to the method described by Rezaei-Sadabady, Eidi,
Zarghami, and Barzegar (2016) with modification. The mixture reac-
tion system was consisted of 100 μL of the test samples at different
concentrations (10, 20, 40, 60, 80 and 100 μg/mL) and 400 μL of a
DPPH-methanol solution (100 μM). The mixture was mixed evenly at
30 °C for 30 min. The absorbance was measured at a wavelength of
517 nm in a SpectraMax Gemini microtiter plate reader (Molecular
Devices, USA). The DPPH scavenging activity of Trolox and ascorbic
acid (Vc) were used as a positive control. The IC50 value was defined as
the amount of total phenolics for a 50% scavenging activity of the free
radicals, which were used to reflect the antioxidant activity. The DPPH
free radical scavenging activity was calculated according to the Eq. (1):
Asample −Ablank ⎞
DPPH radical scavenging activity (%) = ⎛1−
⎜ ⎟ × 100
⎝ Acontrol ⎠ (1)
where Acontrol is the absorbance of DPPH solution, and Asample is the
absorbance of the reacted mixture of DPPH with extract sample. Ablank
is the absorbance of the tested sample with methanol.
2.8.2. ABTS radical cation scavenging activity
The ABTS radical cation (ABTS+) scavenging activity of the samples
was performed according to the method of Tai, Ohno, and Ito (2016).
ABTS+ was generated by oxidation of ABTS (7 mM) with potassium
persulfate (2.45 mM) after incubation in a the dark room at room
temperature for 16 h. The prepared ABTS+ solution was diluted with
ethanol to obtain absorbance at 734 nm of 0.70 ± 0.02 verified by a
spectrophotometer before use. Sample solutions (50 μL) at different
concentrations (5, 10, 20, 30, 40 and 60 μg/mL) of polyphenol were
added to 400 μL of the ABTS+ solution, and the solution was mixed
thoroughly. The Trolox and Vc were used as a positive control. The
reactive mixture was allowed to stand in the dark at room temperature
for 10 min, and the absorbance was then recorded at 734 nm. The
ABTS+ radical scavenging activity was calculated using Eq. (2):
Asample −Ablank ⎞
ABTS+ radical scavenging activity (%) = ⎛1−
⎜ ⎟ × 100
⎝ Acontrol ⎠ (2)
+
where Acontrol is the absorbance of the ABTS solution, Asample is the
absorbance of the reacted mixture of ABTS+ with test sample, and
Ablank is the absorbance of the tested sample and methanol.
2.8.3. Nitrite anion radical (NO2−) scavenging activity
Determination of the NO2 radical-scavenging activity was con-
ducted based on the Griess-Ilosvay reaction method (Singh, Sharma, &
Sharma, 2015), with minor modifications. The reaction mixture con-
sisted of 160 μL sodium nitrite (NaNO2; 10 mg/L) and diluted sample
extracts (5, 10, 25, 40, and 60 μg/mL). The phosphate buffer (0.1 M, pH
7.4) instead of sample extracts was used as the blank. The diluted
Trolox and Vc standards (5, 10, 25, 40, and 60 μg/mL) were used as the
positive control. The above mixtures were incubated at 25 °C for
60 min. After incubation, 80 μL of sulfanilic acid reagent (0.4%, w/w)
was added to the reaction mixture, and the mixture was left standing for
5 min to complete diazotization. Next, 40 μL of 0.2% N-(1-naphthyl)
ethylenediamine dihydrochloride was added and allowed to stand for
30 min. Then, the absorbance was measured spectrophotometrically at
538 nm. The NO2− radical-scavenging activity was calculated as fol-
lows Eq. (3):
L. Wang et al. Food Chemistry 264 (2018) 189–198

NO2− radical - scavenging activity (%) = ⎛1−

Asample −Ablank ⎞
× 100
A 40 c
⎜ ⎟
Soluble phenolic
⎝ Acontrol ⎠ (3) Insoluble-bound phenolic
35

Total phenolic content

b

(mg GAE/g DM)

where Acontrol is the absorbance of the mixtures of sodium nitrite with 30
buffer, Asample is the absorbance of the reacted mixture of sodium nitrite
with test samples, and Ablank is the absorbance of the tested sample and 25
buffer. a
20
a
15
2.8.4. Reducing power
The reducing power of the extracts was determined according to the 10
b b
method of Ferreira, Baptista, Vilas-Boas, and Barros (2007) with slight
5
modifications. A total of 200 μL of each sample with different con-
centrations (20, 40, 60, 120, 160 and 200 μg/mL) was mixed with 0
500 μL of 200 mM of phosphate buffer (pH 6.6) followed by 500 μL of UF F FE
1% potassium ferricyanide. The mixture was incubated for 20 min in a Different processing
water bath at 50 °C. After incubation, 200 μL of 10% trichloroacetic
acid was added, followed by centrifugation at 3000g for 10 min. The
B c
30 Soluble flavonoid
supernatant (500 μL) was mixed with 500 μL of distilled water and Insoluble-bound flavonoid

Total flavonoid content

100 μL of 0.1% ferric chloride; then, the absorbance was read at 700 nm b
25
against a blank. A higher absorbance of the reaction mixture was as-

(mg RE/g DM)

sociated with a greater reducing power. The reducing power was ex- 20
pressed as mmol Trolox equivalents (TE)/g sample in dry mass. The
calibration curves of the standard Trolox were prepared with a standard a
15 a
chemical content between 10 and 200 μg/mL (R2 = 0.9989).
10 b
c
2.9. α-Glucosidase inhibitory assays
5
Determination of α-glucosidase inhibitory activity was based on the
0
method described previously (Hu et al., 2017) with slight modifica-
F

tions. Briefly, the mixture reaction containing 100 μL of α-glucosidase
(1 U/mL) and 100 μL of the extract dilutions (2, 10, 30, 50, 80 and
100 μg/mL) was incubated at 37 °C for 10 min. Instead of the extract
dilutions, 100 μL of phosphate buffer (0.1 M, pH 6.8) was used as the
enzyme control, and 100 μL of phosphate buffer was used instead of the
enzyme solution as the extract control. The acarbose at different con-
centrations were used as the positive control. Next, 100 μL of a p-ni-
trophenyl-α-D-glucopyranoside solution (5 mM) was added to the above
mixture, and the incubation was continued for another 20 min at 37 °C.
The reaction was stopped by adding 500 μL of a 1 M Na2CO3 solution.
The absorbance was measured at 405 nm (A405) using a micro-plate
reader (SpectraMax, M5, Molecular Devices, Sunnyvale, CA, USA). The
IC50, which was defined as the amount of a sample applied to inhibit
50% of the enzyme activity, was used to represent the inhibition effi-
ciency of the extracts against α-glucosidase. The α-glucosidase inhibi-
tion activity (GIA) was calculated as shown in Eq. (4).
(A−B )−(C −D) ⎤
α - glucosidase inhibition activity (%) = ⎡ × 100
⎣ A−B ⎦ (4)
where A, B, C, and D represent the A405 of the enzyme control (con-
taining buffer and α-glucosidase), the blank (containing buffer only),
the reaction mixtures (containing extracts, buffer, and α-glucosidase),
and the extract control (containing extracts and buffer), respectively.
2.10. Statistical analysis
All results presented in this paper were the average of three in-
dependent assays and were expressed as the mean ± standard devia-
tion (SD). The results were analyzed by one-way analysis of variance
(ANOVA). Significant differences were determined by Duncan’s mul-
tiple range tests or by independent sample T-tests when necessary.
Samples with p < 0.05 were considered statistically significant in all
cases. Statistical analyses were conducted using the SPSS version 17.0
(SPSS Inc., Chicago, IL, USA) software package for Windows.
FE
F
U Different processing
Fig. 1. The soluble/insoluble-bound phenolic contents (A) and the soluble/in-
soluble-bound flavonoid contents (B) from GLT after different processing. Each
value represents the mean ± SD (n = 3). GLT, guava leaves tea; UF, un-
fermentation processing; F, fermentation processing; FE, fermentation and
complex enzymatic processing.
3. Results
3.1. Total soluble phenolics and flavonoids
Fig. 1A shows the contents of the total soluble and insoluble-bound
phenolics in GLT following different processing methods. After fer-
mentation treatment, the amounts of total soluble phenolics and in-
soluble-bound phenolics were 31.1 and 5.5 mg GAE/g DM, respectively.
The content of soluble phenolics was enhanced by 71.4% compared to
that of the unfermented GLT, but the content of insoluble-bound phe-
nolics was decreased by 61.3%. Fermentation combined with complex
enzymatic hydrolysis processing further increased the content of total
soluble phenolics (37.9 mg GAE/g DM), which was increased by
108.9% compared to the unfermented stage, and decreased the content
of insoluble-bound phenolics by 65.1%. When compared to fermenta-
tion processing, the soluble phenolics content was enhanced by 21.9%.
After fermentation, complex enzymatic processing resulted in an in-
crease in the ratio of the total soluble phenolic fraction to the total
phenolics from 56.1% to 88.4%, while the ratio of the insoluble-bound
form to total phenolics decreased from 43.9% to 11.6%.
The contents of total soluble and insoluble-bound flavonoids in GLT
following different processing methods are shown in Fig. 1B. Fermen-
tation evidently improved the amount of total soluble flavonoids
(23.3 RE/g DM), which was enhanced by 71.4% compared to that of
unfermented GLT, while the content of insoluble-bound flavonoids was
decreased by 47.1%. After fermentation combined with complex en-
zymatic processing, the total soluble flavonoid content (28.8 mg GAE/

192
L. Wang et al. Food Chemistry 264 (2018) 189–198

4 SMF Fig. 2. HPLC chromatogram of the standard com-

A 2
SMP pounds (A), soluble (B) and insoluble-bound phe-
nolics (C) in GLT after different processings. GLT,
15
3 guava leaves tea; SMF, standard mixtures of flavo-
5
1 noids; SMP, standard mixtures of phenolic acids or L-
6
9 ascorbic acid; SPUF, Soluble phenolics of un-
8 fermented GLT; SPF, Soluble phenolics of fermented
5
GLT; SPFE, Soluble phenolics of fermentation and
complex enzymatic GLT; IBPUF, Insoluble-bound
7 10 11 14 15
12 phenolics of unfermented PGL; IBPF, Insoluble-
13
16 bound phenolics of fermented GLT; IBPFE,
Insoluble-bound phenolics of fermentation and
0 4 8 12 16 20 24 28 32 36 40
complex enzymatic GLT; Peaks: 1, L-ascorbic acid, 2,
Gallic acid, 3, Chlorogenic acid, 4, p-hydroxybenzoic
Time (min)
Soluble phenolics acid, 5, Caffeic acid, 6, Rutin, 7, Isoquercitrin, 8,
1 2
Sinapic acid, 9, Ferulic acid, 10, Quercetin-3-O-β-D-
B xylopyranoside, 11, Quercetin-3-O-α-L-arabinoside,
15 12, Avicularin, 13, Quercitrin, 14, Kaempferol-3-O-
glucose, 15, Quercetin, 16, Kaempferol.

11
3 4 6 16
5 8 9 12 13 14 SPFE
6 10 13
3 4 14 15
5 SPF 16
3 6 11 12
10 13 14 15
4 5 SPUF

0 4 8 12 16 20 24 28 32 36 40
C Time (min)
Insoluble-bound phenolics
12

1 4 11
2 13 IBPFE
3 6 10
5 9
12
1
4 11
2 13
5 6 IBPF
3 910
12
11
1 2 4
6 13 IBPUF
5 89 10
3

0 4 8 12 16 20 24 28 32 36 40
Time (min)

g DM) was further increased by 100.7%, compared to that at the un-
fermented stage. After processing, the ratio of soluble flavonoids to
total flavonoids increased from 51.8% to 81.9%.
3.2. Phenolic compositions
To investigate the changes in the compositions and the contents of
the soluble and insoluble-bound phenolics from GLT with different
processing methods, the quantities of 16 compounds, including gallic
only in the soluble form and rarely existed in the insoluble-bound form.
After fermentation processing, the soluble contents of most of the
phenolics from the GLT extracts were slightly increased (p = 0.002),
except for the syringic acid and ferric acid contents, which were dis-
tinctly decreased (p < 0.05). The increases in the total solubility of
each phenolic compound following fermentation treatment were as
follows: gallic acid, 14.3%; p-hydroxybenzoic acid, 34.9%; rutin,
214.6%; p-coumaric acid, 46.4%; isoquercitrin, 26.6%; quercetin-3-O-β-
D-xylopyranoside, 73.4%; quercetin-3-O-α-L-arabinoside, 60.1%; quer-

acid, chlorogenic acid, p-coumaric acid, sinapic acid, p-hydroxybenzoic
acid, ferulic acid, caffeic acid, rutin, quercetin-3-O-β-D-xylopyranoside,
quercetin-3-O-α-L-arabinoside, isoquercitrin, quercitrin, avicularin,
kaempferol-3-O-glucose, quercetin and kaempferol, were measured.
The results are shown in Fig. 2A–C and Table 1. According to the HPLC
analyses, the compositions of soluble and insoluble-bound phenolics
were similar after different processing treatments, but their contents
were evidently different. From Fig. 2A–C and Table 1, most of the de-
termined phenolics were present in the soluble and insoluble-bound
forms, except chlorogenic acid and kaempferol, which were present
citrin, 26.9%; quercetin, 472.0%; and kaempferol, 236.9%. When
treated with complex enzymatic processing after fermentation, the
compositions of soluble phenolic fractions were significantly changed.
Among them, isoquercitrin, quercetin-3-O-β-D-xylopyranoside, and
avicularin were evidently decreased and were almost completely
transferred into quercetin, and kaempferol-3-O-glucose was transferred
into kaempferol. Compared to the unfermented GLT, the increases in
the soluble form of individual phenolic compound in GLT after fer-
mentation combined with complex enzymatic treatment were as fol-
lows: gallic acid, 32.9%; chlorogenic acid, 25.1%; p-coumaric acid,

193
L. Wang et al. Food Chemistry 264 (2018) 189–198

Table 1 Table 2
Changes in individual phenolics of the soluble and insoluble bound phenolics Antioxidant activity, reducing power and the inhibition activity of α-glucosi-
fractions in GLT following different processing. dase of the soluble and insoluble-bound phenolics fractions of guava leaves tea
extracts at different processing stages.
Analytes Stage Soluble phenolics Insoluble-bound
(mg/100 g DM) (mg/100 g DM) Stages Samples Control

Gallic acid UF 147.3 ± 1.3Ba 71.4 ± 2.1Ac Soluble Insoluble-bound Vc Trolox

F 168.5 ± 0.7Bb 53.9 ± 0.3Ab
FE 195.7 ± 0.8Bc 37.3 ± 0.2Aa +
IC50 for scavenging effects of ABTS
(μg/mL)
Chlorogenic acid UF 24.2 ± 0.4Ba 3.2 ± 0.2Ab
UF 18.6 ± 0.2Cc 27.8 ± 1.3 Da
F 28.6 ± 0.1b N.D.c
F 9.4 ± 0.3Bb 33.2 ± 2.2Db 18.3 ± 0.9Ca 12.3 ± 0.1Ba
FE 30.3 ± 0.8b N.D.
FE 4.5 ± 0.3Aa 36.2 ± 1.1Db
p-Hydroxybenzoic acid UF 10.1 ± 0.1Ba 3.2 ± 0.1Ab
IC50 for scavenging effects of DPPH
F 13.6 ± 0.01Bb 3.0 ± 0.1Ab
(μg/mL)
FE 13.7 ± 0.01Bb 2.7 ± 0.01Aa
UF 39.5 ± 1.3Cc 60.3 ± 2.6 Da
Caffeic acid UF 5.6 ± 0.1Aa 4.1 ± 0.01Ab F 27.1 ± 0.9Bb 75.2 ± 3.2Db 38.6 ± 1.0Cc 39.6 ± 0.5Cc
F 6.7 ± 0.3Bb 3.0 ± 0.01Ab FE 14.7 ± 0.3Aa 80.7 ± 2.7Dc
FE 7.0 ± 0.12Bb 1.5 ± 0.02Aa
IC50 for scavenging effects of NO2−
Rutin UF 1.5 ± 0.01Aa 2.2 ± 0.07Bc (μg/mL)
F 48.1 ± 0.04Bb 0.3 ± 0.05Ab UF 14.3 ± 1.0Ac 20.5 ± 1.8Ba
FE 1.5 ± 0.01Ba 0.1 ± 0.01Aa F 8.5 ± 0.5Ab 33.0 ± 1.1Cb 21.4 ± 1.1Bc 17.8 ± 0.1Ba
FE 2.8 ± 0.2Aa 38.1 ± 2.0Dc
p-Coumaric acid UF 9.2 ± 0.05Aa 17.7 ± 0.8Bc
F 13.5 ± 0.07Ab 11.4 ± 0.08Ab Reducing power (mmol TE/g DM)
FE 16.5 ± 0.07Bc 9.3 ± 0.05Aa UF 7.4 ± 1.6Aa 11.4 ± 0.3Bb
F 50.3 ± 2.3Bb 3.5 ± 0.1Aa
Isoquercitrin UF 19.6 ± 0.26Ab 22.5 ± 0.21Ab
FE 68.9 ± 1.6Bc 2.9 ± 0.6Aa
F 24.8 ± 0.35Bc 7.2 ± 0.01Aa
FE 6.8 ± 0.15Aa 5.3 ± 0.04Aa IC50 for the inhibition activity of α- Acarbose
glucosidase (μg/mL)
Sinapic acid UF 9.2 ± 0.07Bc 6.3 ± 0.03Ab
UF 29.1 ± 1.5Ac 71.6 ± 2.3Ba
F 6.3 ± 0.04Ab 5.3 ± 0.01Aa
F 19.2 ± 1.3Ab 104.4 ± 1.9Bb 178.52 ± 3.1Ba
FE 2.3 ± 0.01Aa 4.2 ± 0.06Ba
FE 11.8 ± 1.6Aa 112.2 ± 2.1Bb
Ferulic acid UF 12.9 ± 0.03Ba 9.4 ± 0.11Ab
F 13.9 ± 0.07Ba 7.1 ± 0.2Aa UF, Unfermentation processing; F, Fermentation processing; FE, Fermentation
FE 12.4 ± 0.01Ba 6.3 ± 0.01Aa and complex enzyme hydrolysis processing. IC50 values were defined as the
Quercetin-3-O-β-D- UF 18.3 ± 0.2Bb 11.3 ± 0.2Ac amount of total phenolics for a 50% scavenging activity of the free radicals.
xylopyranoside F 31.8 ± 0.1Bc 7.4 ± 0.01Ab Each value was expressed as mean ± standard deviation (n = 3). Values with
FE 1.5 ± 0.08Ba 3.5 ± 0.02Aa different letters (within row in uppercase letters (A–D), within columns in
Quercetin-3-O-α-L- UF 42.3 ± 0.2Ba 15.4 ± 0.4Ac lowercase letters (a–c)) are significantly different (p < 0.05).
arabinopyranoside F 67.8 ± 0.2Bc 10.4 ± 0.3Ab
FE 55.7 ± 0.07Bb 7.5 ± 1.1Aa hydroxybenzoic acid, caffeic acid, and ferulic acid.
Avicularin UF 57.7 ± 0.2Bb 41.4 ± 0.2Ac
F 60.3 ± 0.2Bb 32.3 ± 0.2Ab
FE 2.3 ± 0.02Aa 4.1 ± 0.1Ba 3.3. Antioxidant activities and reducing power
Kaempferol-3-O-glucoside UF 14.2 ± 0.07Bb 8.3 ± 0.03Ac
F 18.1 ± 0.2Bc 3.5 ± 0.02Ab To completely reflect the antioxidant capacity of samples, three
FE 1.2 ± 0.06Aa 1.1 ± 0.07Aa
anti-oxidant modes including DPPH, ABTS+ , and NO2− free radical
Quercitrin UF 22.7 ± 0.2Bb 17.3 ± 0.01Ac scavenging activity were used to evaluate the anti-oxidant capacity of
F 28.8 ± 0.2Bc 11.6 ± 0.06Ab
the soluble and insoluble-bound phenolic extracts in GLT following
FE 11.3 ± 0.2Ba 8.4 ± 0.07Ab
different processing methods. The results are shown in Fig. S1A–F and
Quercetin UF 24.1 ± 0.2Aa 34.2 ± 0.05Bb Table 2. SPUF showed higher DPPH radical scavenging activity
F 141.6 ± 0.3Bb 2.5 ± 0.09Aa
FE 312.8 ± 1.2c N.D.
(IC50 = 39.5 μg/mL), and it showed no significant differences from the
positive control Trolox (IC50 = 39.5 μg/mL) and Vc (IC50 = 38.6 μg/
Kaempferol UF 3.1 ± 0.1Aa 4.2 ± 0.03A
mL). However, SPFE showed higher DPPH radical scavenging activity
F 10.5 ± 0.5b N.D.
FE 21.3 ± 0.4c N.D. than SPUF and SPF, as revealed in Fig. S1A. With the concentrations of
SPFE at 20 μg/mL and 40 μg/mL, the DPPH radical scavenging activity
UF, unfermentation processing; F, fermentation processing; FE, fermentation of SPFE reached up to 58.6% and 81.2%, respectively. However, the
and complex enzyme hydrolysis processing. Each value was expressed as activity of IBPFE was at only 11.4% and 26.9% at the same con-
mean ± standard deviation (n = 3). Values with different letters (within row centrations (Fig. S1B). The DPPH radical scavenging activity of SPFE
in uppercase letters (A–B), within columns in lowercase letters (a–c)) are sig- (IC50 = 14.7 μg/mL) was apparently higher than those of SPUF
nificantly different (p < 0.05). N.D. Not detected.
(IC50 = 39.5 μg/mL) and SPF (IC50 = 27.1 μg/mL) (p < 0.01).
The results shown in Fig. S1C also confirmed that SPFE had higher
79.1%; quercetin-3-O-α-L-arabinoside, 31.8%; quercetin, 1163.6%; and
ABTS+ radical scavenging activity than SPUF and SPF. Moreover, the
kaempferol, 581.1%. The five phenolic compounds that showed de-
ABTS+ radical scavenging activity of SPFE and SPF were also sig-
creases in the soluble form were syringic acid, 74.8%; isoquercitrin,
nificantly higher than that of the positive control Trolox. When the
65.5%; quercitrin, 50.3%; quercetin-3-O-β-D-xylopyranoside, 91.7%,
concentration of SPFE was 20 μg/mL, the ABTS+ radical scavenging
and avicularin 96.1%. Compared to fermented GLT, the three phenolic
activity reached 81.7%. For SPF and SPUF, the ABTS+ radical
compounds that showed increases in the soluble form were gallic acid,
scavenging activities were only 54.1% and 69.4%, respectively. How-
16.1%; quercetin, 120.9%; and kaempferol, 102.86%. But there were
ever, the activity of IBPFE was at only 29.7% at the same concentration
no changes in the contents of the following individual phenolics: p-
(Fig. S1D). The ABTS+ radical scavenging activity of SPFE

194
L. Wang et al. Food Chemistry 264 (2018) 189–198

A 1.0 A
Trolox
SPUF 100 SPUF
Reducing power (A700nm)

Į-Glucosidase inhibition activity (%)

0.8 SPF
SPF
SPFE
SPFE
80
0.6

60
0.4

40
0.2

20
0.0
20 40 60 80 100 120 140 160 180 200
Sample concentration (ȝg/mL) 0
B 0 10 20 30 40 50 60 70 80 90 100
Sample concentrations (ȝg/mL)
0.5 IBPUF
IBPF
B
70
Reducing power (A700nm)

IBPFE

Į-Glucosidase inhibition activity (%)

IBPUF
0.4
60 IBPF
IBPFE

0.3 50

40
0.2
30
0.1 20

10
0.0
20 40 60 80 100 120 140 160 180 200
0
Sample concentration (ȝg/mL) 0 10 20 30 40 50 60 70 80 90 100
Fig. 3. Reducing power of soluble (B) and insoluble-bound (B) phenolics ex- Sample concentrations (ȝg/mL)
tracts at different concentrations from GLT after different processing. Each C
value represents the mean ± SD (n = 3). GLT, guava leaves tea; SPUF, Soluble 60
Į-Glucosidase inhibitory potency (%)

phenolics of unfermented GLT; SPF, Soluble phenolics of fermented GLT; SPFE,

Soluble phenolics of fermentation and complex enzymatic GLT; IBPUF, 50
Insoluble-bound phenolics of unfermented GLT; IBPF, Insoluble-bound phe-
nolics of fermented GLT; IBPFE, Insoluble-bound phenolics of fermentation and
complex enzymatic GLT.
40

(IC50 = 4.5 μg/mL) was also apparently higher than those of SPF
(IC50 = 9.4 μg/mL) (p < 0.01), SPUF (IC50 = 18.6 μg/mL), Trolox
(IC50 = 18.3 μg/mL) and Vc (IC50 = 12.3 μg/mL).
The NO2− radical scavenging activity of the soluble and insoluble-
bound phenolics in GLT with different processing methods are shown in
Fig. S1E, F and Table 2. Similarly, we found that SPUF showed strong
NO2− radical scavenging activity, which was significantly higher than
that of the positive control Trolox. Importantly, the NO2− radical
scavenging activity of SPFE and SPF were also evidently higher than
that of SPUF. The NO2− radical scavenging activity of SPF and SPUF
were 79.1% and 62.3%, respectively, when the concentration was
20 μg/mL. For SPFE, the scavenging activity reached 89.5% (Fig. S1E).
However, the activity of IBPFE was at only 29.72% at the same con-
centration (Fig. S1F). The NO2− radical scavenging activity of SPFE
(IC50 = 2.8 μg/mL) was also distinctly higher than those of SPF
(IC50 = 8.5 μg/mL), SPUF (IC50 = 14.3 μg/mL), Trolox
(IC50 = 17.8 μg/mL) and Vc (IC50 = 21.4 μg/mL) (p < 0.01).
Fig. 3A, B and Table 2 show the reducing power of the soluble and
insoluble-bound phenolics in GLT treated with different processing
methods. The reducing power of all sample extracts and Trolox in-
creased with increasing concentrations. Compared to other samples,
SPFE displayed the highest the reducing power. When the
30
20
10
0
20 40 60 80 100 120 140 160 180 200
Acarbose concentrations (ȝg/mL)
Fig. 4. The α-glucosidase inhibition acvtivities of soluble (A), insoluble-bound
phenolics (B) extracts from GLT after different processing and acarbose (C).
Each value represents the mean ± SD (n = 3). GLT, guava leaves tea; SPUF,
Soluble phenolics of unfermented GLT; SPF, Soluble phenolics of fermented
GLT; SPFE, Soluble phenolics of fermentation and complex enzymatic GLT;
IBPUF, Insoluble-bound phenolics of unfermented GLT; IBPF, Insoluble-bound
phenolics of fermented GLT; IBPFE, Insoluble-bound phenolics of fermentation
and complex enzymatic GLT.
concentration was 200 μg/mL, the reducing power reached
68.6 mmol TE/g DM (Fig. 3A). The reducing power was on the order of
SPFE (97.86 mmol TE/g DM) > SPF (50.3 mmol TE/g DM) > SPUF
(7.4 mmol TE/g DM). Interestingly, the reducing power of SPFE was

195
L. Wang et al.
also evidently much higher than those of SPF and SPUF (p < 0.01).
Moreover, the reducing power of soluble phenolics (SPF/SPFE) was also
much higher than that of the insoluble-bound phenolics (IBF/IBFE) at
the same concentration (Fig. 3B). The reducing power of IBPUF
(11.4 mmol TE/g DM) was higher than those of IBFE (3.5 mmol TE/
g DM) and IBF (2.9 mmol TE/g DM).
3.4. The inhibition activity of α-glucosidase
As shown in Fig. 4A, B and Table 2, the inhibition activity of α-
glucosidase for all sample extracts also increased with the concentration
of phenolics. SPUF showed notable higher inhibitory efficiency against
α-glucosidase. However, the inhibitory efficacy of α-glucosidase was
evidently increased after fermentation. When the concentration was
30 μg/mL, the inhibition activity of α-glucosidase reached 72%
(Fig. 4A). More importantly, the inhibition activity of α-glucosidase of
SPFE (IC50 = 11.8 μg/mL) was also distinctly higher than those of SPF
(IC50 = 19.2 μg/mL) and SPUF (IC50 = 29.1 μg/mL) (p < 0.01).
Moreover, soluble phenolics showed higher inhibitory potency with α-
glucosidase than insoluble-bound phenolics. The inhibition activity of
IBPF (IC50 = 104.4 μg/mL) and IBPFE (IC50 = 112.2 μg/mL)
(p < 0.05) were distinctly lower than that of IBPUF (IC50 = 71.6 μg/
mL) (Fig. 4 and Table 2). From Fig. 4C, The α-glucosidase inhibition
activity of acarbose (positive control) (IC50 = 178.52 μg/mL) was also
distinctly lower than those of all samples extracts. This also explains
guava leaf tea have a good application in the treatment of hypergly-
cemia in the fork.
4. Discussion
4.1. Effects of different processing on the total soluble phenolics and soluble
flavonoids
The results showed SSF processing significantly improved the re-
leasing of soluble phenolics from GLT. It is due to that co-fermentation
with M. anka and S. cerevisiae could utilize sugars or polysaccharides to
produce cellulases, ligninase and feruloyl esterases, which can hydro-
lyze some chemical bonds between phenolics and cell wall structural
elements to release the soluble phenolics (Alshikh, de Camargo, &
Shahidi, 2015; Wang, Geng, Egashira, & Sanada, 2004). Zhang et al.
(2017) reported that SSF with Aspergillus niger enhanced phenolic
contents and their antioxidant activities. In this study, complex enzy-
matic processing after fermentation was conducted to further release
the soluble phenolic compounds and to obtain flavonoid aglycones
(quercetin and kaempferol) with stronger bioactivities from GLT. The
results revealed that complex enzymatic processing distinctly increased
the soluble phenolic content when compared to fermentation proces-
sing. It may be due to the complexity of plant cell structures and the
limitation of specific enzymes produced by M. anka and S. cerevisiae,
fermentation combined with enzymatic hydrolysis had an efficient ef-
fect on improving the contents of soluble phenolics. Liu et al. (2017)
reported that fermentation with two lactic acid bacteria improved the
soluble phenolic content from rice bran pretreated with α-amylase. In
the study, due to the complexity of plant cell wall structural elements
and the characteristics of phenolics compositions (mainly flavonoids
glycosides) from GLT, complex enzymes consisted of cellulase, β-glu-
cosidase, xylanase and hemicellulase were selected to carry out the
study. These enzymes can not only further release the soluble phenolics
from fermented GLT but also greatly convert the flavonoid glycosides
into flavonoid aglycones with stronger biological activities.
4.2. Effect of different processing on the phenolic composition
The changes in individual phenolics in the soluble and insoluble
bound forms from GLT following different processing were in-
vestigated. We found the compositions of the main phenolics from GLT
196
Food Chemistry 264 (2018) 189–198
consisted of phenolic acids (gallic acid, chlorogenic acid, p-coumaric
acid, sinapic acid, p-hydroxybenzoic acid, ferulic acid, caffeic acid,
quercetin and kaempferol) and flavonoid glycoside compounds (rutin,
quercetin-3-O-β-D-xylopyranoside, quercetin-3-O-α-L-arabinoside, iso-
quercitrin, quercitrin, avicularin, and kaempferol-3-O-glucose).
Moreover, most of the phenolic compounds were present in the soluble
and insoluble-bound forms. Generally, after fermentation, the contents
of most of the individual phenolics in the soluble form were sig-
nificantly enhanced due to the release of the insoluble-bound form,
expect for sinapic acid. Liu et al. (2017) reported decreases in the si-
napic acid content may be due to degradation upon exposure to high
temperatures. Moreover, fermentation processing released 141.6 mg/
100 g DM of quercetin and 10.5 mg/100 g DM of kaempferol from GLT,
which were 5.9 and 3.4 times that of the unfermented control, re-
spectively. This result was due to the microorganisms producing some
key enzymes that can damage the cell wall to promote the release of
soluble phenolic compounds. The results were consistent with those
from previous reports in which fermentation increased some phenolic
and flavonoid compounds from rice bran and tartary buckwheat leaves
(Liu et al., 2017; Zhang et al., 2017). More interestingly, complex en-
zymatic processing after fermentation not only further increased the
individual phenolic contents but also greatly enhances the quercetin
and kaempferol contents. Compared to the unfermented control, the
contents were increased 13.0 times and 6.8 times, respectively. This
increase may be due to co-fermentation with M. anka and S. cerevisiae,
which first improved the release of soluble phenolics, with most of the
phenolics mainly existing in the soluble form. Then, complex enzymes,
especially for β-glucosidase, can better play a role in maximally con-
verting the flavonoids glycosides (isoquercitrin, quercitrin, quercetin-3-
O-β-D-xylopyranoside, kaempferol-3-O-glucose and avicularin) into
quercetin and kaempferol (Day et al., 1998). Liu et al. (2017) confirmed
complex enzymatic hydrolysis processing after fermentation with lactic
acid bacteria strains significantly improved the release of free phenolic
and conjugated phenolics from rice bran. Mathew and Abraham (2004)
also reported that complex enzymes including cellulases, β-glucosidase
and some esterases can promote the release of soluble phenolic com-
pounds from wheat bran. Moreover, many studies have confirmed that
complex enzymatic have obviously higher efficient than a single en-
zymatic to improve the releasing of phenolics compounds from food
matrices (Liu et al., 2017; Mathew & Abraham, 2004).
4.3. Effect of different processing on the antioxidant activity and reducing
power
Phenolics are important natural antioxidants that exist in plant
materials, tea-leaf products and cereal food (Bei et al., 2017; Wang
et al., 2017). It is difficult for a single assay to accurately reflect all the
individual antioxidants in samples (Jessica et al., 2009; Shahidi &
Zhong, 2015). Thus, multiple assay methods were used to evaluate the
antioxidant activity of soluble and insoluble phenolics from GLT pro-
cessed by different methods, through measuring DPPH, ABTS+, NO2−
radical scavenging activity assays and the reducing power. The results
confirmed that fermentation processing significantly increased the
DPPH, ABTS+ and NO2− radical scavenging activities and reducing
power of soluble phenolics. Moreover, the antioxidant activity and re-
ducing power assays obtained similar results, which were consistent
with the changes in the phenolic contents. Fermentation and complex
enzyme processing further promoted the releasing of soluble phenolics
from GLT and resulted in a decrease in the content of insoluble-bound
phenolics. Importantly, complex enzyme processing after fermentation
further improved the antioxidant activity and reducing power of the
soluble phenolics. Many reports confirmed that quercetin and kaemp-
ferol possessed much higher antioxidant activity than their glycoside
compounds (Nuutila, Puupponen-Pimiä, Aarni, & Oksman-Caldentey,
2003; Tabart, Kevers, Pincemail, Defraigne, & Dommes, 2009). After
complex enzyme processing, flavonoid glycosides (isoquercitrin,
L. Wang et al.
quercitrin, quercetin-3-O-β-D-xylopyranoside, kaempferol-3-O-glucose
and avicularin) were transformed into quercetin and kaempferol.
Consequently, the antioxidant activity and reducing power of the so-
luble phenolics in GLT treated with complex enzyme processing were
much higher than those of the unfermented or fermented GLT. Mean-
while, the antioxidant capacity determined by an ABTS assay was sig-
nificantly and consistently higher than the antioxidant capacity de-
termined by a DPPH assay. Floegel, Kim, Chung, Koo, and Chun (2011)
reported the ABTS assay is based on the generation of a blue/green
ABTS+, which is applicable to both hydrophilic and lipophilic anti-
oxidant systems, whereas the DPPH assay uses a radical dissolved in
organic media and is applicable to hydrophobic systems. Hence, ABTS
is more active than DPPH in antioxidant activity assays for phenolics in
GLT extracts.
Our results confirmed that soluble phenolics from GLT extracts were
the main contributors to antioxidant activity and reducing power. So,
fermentation with complex enzymatic is a good strategy for improving
the soluble phenolic contents by releasing the insoluble-bound phe-
nolics in GLT. Ti, Zhang, Li, Wei, and Zhang (2015) reported a large
amount of bound antioxidative phenolics located in the cell wall or
aleurone layer was released with enzymatic treatment. Liu et al. (2017)
reported multiple enzyme hydrolysis changed the composition of phe-
nolics and amplified antioxidant activity in the form of phenolic acids,
such as gallic acid and quercetin. To summarize, there are two methods
for improving the antioxidant activity and reducing power: one was to
increase the solubility of the phenolic compound contents from GLT
extracts and the other was to obtain more phenolic compounds with
higher bioactivities such as quercetin and kaempferol using the bio-
transformation method.
4.4. Effect of different processing on the inhibition activity of α-glucosidase
Though many different medicines are available to manage diabetes,
some side effects are generated due to their application. Studies to
identify alternative α-glucosidase inhibitors with fewer side effects have
been conducted. Phenolic compounds have shown strong inhibitory
activity against α-glucosidase, which is a key enzyme regulating the
absorption of glucose in the small intestine (Eid & Haddad, 2017;
Ranilla et al., 2010). In the study, fermentation promoted the release of
soluble phenolics from GLT, which therefore resulted in better inhibi-
tion on α-glucosidase. Some reports have confirmed phenolic acids and
flavonoids from black legumes or green tea extracts possessed stronger
inhibition effects towards α-glucosidase (Oki, Matsui, & Osajima, 1999;
Tan, Chang, & Zhang, 2017). Wang, Wei, Tian, Shi, and Wu (2016) have
reported fermentation with Lactobacillus plantarum to improve the α-
glucosidase inhibitory effect in vitro and ameliorate blood glucose in
mice for purple Jerusalem artichoke extract. However, complex enzyme
hydrolysis after fermentation processing not only further improved the
content of soluble phenolics from GLT but also greatly increased the
quercetin and kaempferol contents. As we known, quercetin can in-
teract with many molecular targets in the small intestine, liver, skeletal
muscle, pancreas, and adipose tissue and can regulate and maintain
blood glucose homeostasis in the human body (Eid & Haddad, 2017).
The regulation mechanisms of quercetin included the inhibition of
glucose absorption in the intestines, insulin-sensitizing activities, in-
sulin secretion and improved glucose utilization in peripheral tissues.
Moreover, the absorption efficiency in human intestines and the in-
hibitory potential towards α-glucosidase of quercetin and kaempferol
were much higher than those of the glycoside derivatives (Manach
et al., 1997). As mentioned before, α-glucosidase plays an important
role in the absorption of glucose; therefore, the increased levels of so-
luble-phenolics, quercetin and kaempferol in GLT could enhance in-
hibitory activity against α-glucosidase.
197
Food Chemistry 264 (2018) 189–198
5. Conclusions
In conclusion, a large amount of phenolics was released in the so-
luble form from GLT co-fermented with M. anka and S. cerevisiae.
Furthermore, complex enzyme hydrolysis after fermentation not only
further improved the contents of total soluble phenolics, but also
changed the composition of the soluble phenolics. Compared to the
unfermented control, the contents of soluble phenolics, flavonoids,
quercetin and kaempferol were significantly increased. Quercetin was
released in the highest amount among different phenolics. Additionally,
soluble phenolic extracts from GLT following fermentation and complex
enzymatic hydrolysis processing exhibited much higher antioxidant
activities, reducing power and inhibitory efficacy towards α-glucosi-
dase. Our study also provided useful information for processing GLT
that are rich in soluble phenolics and flavonoid aglycones with up-
graded bio-activities.
6. Conflict of interest statement
The authors declared no conflict of interest.
Acknowledgements
The work was supported by the Science and Technology Project of
Guangdong Province, China (2016A020210011 and 2017B020207003)
and the Special fund for Agricultural Science and Technology Research
Project of Jiangmen City, China (20150160008347).
References
Albishi, T., John, J. A., Al-Khalifa, A. S., & Shahidi, F. (2013). Antioxidant, anti-in-
flammatory and DNA scission inhibitory activities of phenolic compounds in selected
onion and potato varieties. Journal of Functional Foods, 5(2), 930–939.
Alshikh, N., de Camargo, A. C., & Shahidi, F. (2015). Phenolics of selected lentil cultivars:
Antioxidant activities and inhibition of low-density lipoprotein and DNA damage.
Journal of Functional Foods, 18, 1022–1038.
Bei, Q., Liu, Y., Wang, L., Chen, G., & Wu, Z. (2017). Improving free, conjugated, and
bound phenolic fractions in fermented oats (Avena sativa L.) with Monascus anka and
their antioxidant activity. Journal of Functional Foods, 32, 185–194.
Bhandari, M. R., Jong-Anurakkun, N., Hong, G., & Kawabata, J. (2008). α-Glucosidase
and α-amylase inhibitory activities of Nepalese medicinal herb Pakhanbhed (Bergenia
ciliata, Haw.). Food Chemistry, 106(1), 247–252.
Cai, S., Wang, O., Wu, W., Zhu, S., Zhou, F., Ji, B., & Cheng, Q. (2011). Comparative study
of the effects of solid-state fermentation with three filamentous fungi on the total
phenolics content (TPC), flavonoids, and antioxidant activities of subfractions from
oats (Avena sativa L.). Journal of Agricultural and Food Chemistry, 60(1), 507–513.
de Araújo, M. E. M. B., Franco, Y. E. M., Alberto, T. G., Sobreiro, M. A., Conrado, M. A.,
Priolli, D. G., & de Oliveira Carvalho, P. (2013). Enzymatic de-glycosylation of rutin
improves its antioxidant and antiproliferative activities. Food Chemistry, 141(1),
266–273.
de Camargo, A. C., Regitano-d’Arce, M. A. B., Biasoto, A. C. T., & Shahidi, F. (2016).
Enzyme-assisted extraction of phenolics from winemaking by-products: Antioxidant
potential and inhibition of alpha-glucosidase and lipase activities. Food Chemistry,
212, 395–402.
Day, A. J., DuPont, M. S., Ridley, S., Rhodes, M., Rhodes, M. J., Morgan, M. R., &
Williamson, G. (1998). Deglycosylation of flavonoid and isoflavonoid glycosides by
human small intestine and liver β-glucosidase activity. FEBS Letters, 436(1), 71–75.
Dey, T. B., Chakraborty, S., Jain, K. K., Sharma, A., & Kuhad, R. C. (2016). Antioxidant
phenolics and their microbial production by submerged and solid state fermentation
process: A review. Trends in Food Science & Technology, 53, 60–74.
Díaz-de-Cerio, E., Gómez-Caravaca, A. M., Verardo, V., Fernández-Gutiérrez, A., &
Segura-Carretero, A. (2016). Determination of guava (Psidium guajava L.) leaf phe-
nolic compounds using HPLC-DAD-QTOF-MS. Journal of Functional Foods, 22,
376–388.
Dikshit, R., & Tallapragada, P. (2016). Statistical optimization of lovastatin and con-
firmation of nonexistence of citrinin under solid-state fermentation by Monascus
sanguineus. Journal of Food and Drug Analysis, 24(2), 433–440.
Dulf, F. V., Vodnar, D. C., Dulf, E. H., & Tosa, M. I. (2015). Total phenolic contents,
antioxidant activities, and lipid fractions from berry pomaces obtained by solid-state
fermentation of two Sambucus species with Aspergillus niger. Journal of Agricultural
and Food Chemistry, 63(13), 3489–3500.
Dulf, F. V., Vodnar, D. C., & Socaciu, C. (2016). Effects of solid-state fermentation with
two filamentous fungi on the total phenolic contents, flavonoids, antioxidant activ-
ities and lipid fractions of plum fruit (Prunus domestica L.) by-products. Food
Chemistry, 209, 27–36.
Eid, H., & Haddad, P. S. (2017). The antidiabetic potential of quercetin: Underlying
L. Wang et al.
mechanisms. Current Medicinal Chemistry, 24(4), 355–364.
Ferreira, I. C., Baptista, P., Vilas-Boas, M., & Barros, L. (2007). Free-radical scavenging
capacity and reducing power of wild edible mushrooms from northeast Portugal:
Individual cap and stipe activity. Food Chemistry, 100(4), 1511–1516.
Floegel, A., Kim, D. O., Chung, S. J., Koo, S. I., & Chun, O. K. (2011). Comparison of
ABTS/DPPH assays to measure antioxidant capacity in popular antioxidant-rich US
foods. Journal of Food Composition and Analysis, 24(7), 1043–1048.
Hu, P., Li, D. H., Jia, C. C., Liu, Q., Wang, X. F., Li, Z. L., & Hua, H. M. (2017). Bioactive
constituents from Vitex negundo var. heterophylla and their antioxidant and α-glu-
cosidase inhibitory activities. Journal of Functional Foods, 35, 236–244.
Liu, L., Zhang, R., Deng, Y., Zhang, Y., Xiao, J., Huang, F., ... Zhang, M. (2017).
Fermentation and complex enzyme hydrolysis enhance total phenolics and anti-
oxidant activity of aqueous solution from rice bran pretreated by steaming with α-
amylase. Food Chemistry, 221, 636–643.
Manach, C., Morand, C., Demigné, C., Texier, O., Régérat, F., & Rémésy, C. (1997).
Bioavailability of rutin and quercetin in rats. FEBS Letters, 409(1), 12–16.
Mathew, S., & Abraham, T. E. (2004). Ferulic acid: An antioxidant found naturally in
plant cell walls and feruloyl esterases involved in its release and their applications.
Critical Reviews in Biotechnology, 24(2–3), 59–83.
Nuutila, A. M., Puupponen-Pimiä, R., Aarni, M., & Oksman-Caldentey, K. M. (2003).
Comparison of antioxidant activities of onion and garlic extracts by inhibition of lipid
peroxidation and radical scavenging activity. Food Chemistry, 81(4), 485–493.
Oki, T., Matsui, T., & Osajima, Y. (1999). Inhibitory effect of α-glucosidase inhibitors
varies according to its origin. Journal of Agricultural and Food Chemistry, 47(2),
550–553.
Qiu, Y., Liu, Q., & Beta, T. (2010). Antioxidant properties of commercial wild rice an-
danalysis of soluble and insoluble phenolic acids. Food Chemistry, 121(1), 140–147.
Ranilla, L. G., Kwon, Y. I., Apostolidis, E., & Shetty, K. (2010). Phenolic compounds,
antioxidant activity and in vitro inhibitory potential against key enzymes relevant for
hyperglycemia and hypertension of commonly used medicinal plants, herbs and
spices in Latin America. Bioresource Technology, 101(12), 4676–4689.
Rezaei-Sadabady, R., Eidi, A., Zarghami, N., & Barzegar, A. (2016). Intracellular ROS
protection efficiency and free radical-scavenging activity of quercetin and quercetin-
encapsulated liposomes. Artificial Cells, Nanomedicine, and Biotechnology, 44(1),
128–134.
Shahidi, F., & Zhong, Y. (2015). Measurement of antioxidant activity. Journal of
Functional Foods, 18, 757–781.
Singh, R., Sharma, S., & Sharma, V. (2015). Comparative and quantitative analysis of
antioxidant and scavenging potential of Indigofera tinctoria Linn. extracts. Journal of
Integrative Medicine, 13(4), 269–278.
198
Food Chemistry 264 (2018) 189–198
Singleton, V. L., Orthofer, R., & Lamuela-Raventos, R. M. (1999). Analysis of total phenols
and other oxidation substrates and antioxidants by means of Folin Ciocalteu reagent.
Methods in Enzymology, 299, 152–178.
Tabart, J., Kevers, C., Pincemail, J., Defraigne, J. O., & Dommes, J. (2009). Comparative
antioxidant capacities of phenolic compounds measured by various tests. Food
Chemistry, 113(4), 1226–1233.
Jessica, T., Claire, K., Joël, P., Jeanolivier, D., & Jacques, D. (2009). Comparative anti-
oxidant capacities of phenolic compounds measured by various tests. Food Chemistry,
113(4), 1226–1233.
Tai, A., Ohno, A., & Ito, H. (2016). Isolation and characterization of the 2,2′-azinobis (3-
ethylbenzothiazoline-6-sulfonic acid) (ABTS) radical cation-scavenging reaction
products of arbutin. Journal of Agricultural and Food Chemistry, 64(38), 7285–7290.
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.
Ti, H., Zhang, R., Li, Q., Wei, Z., & Zhang, M. (2015). Effects of cooking and in vitro
digestion of rice on phenolic profiles and antioxidant activity. Food Research
International, 76, 813–820.
Van Hung, P. (2016). Phenolic compounds of cereals and their antioxidant capacity.
Critical Reviews in Food Science and Nutrition, 56(1), 25–35.
Wang, L., Bei, Q., Wu, Y., Liao, W., & Wu, Z. (2017). Characterization of soluble and
insoluble-bound polyphenols from Psidium guajava L. leaves co-fermented with
Monascus anka and Bacillus sp. and their bio-activities. Journal of Functional Foods, 32,
149–159.
Wang, L., Wei, W., Tian, X., Shi, K., & Wu, Z. (2016a). Improving bioactivities of poly-
phenol extracts from Psidium guajava L. leaves through co-fermentation of Monascus
anka GIM 3.592 and Saccharomyces cerevisiae GIM 2.139. Industrial Crops and
Products, 94, 206–215.
Wang, Z., Hwang, S. H., Lee, S. Y., & Lim, S. S. (2016b). Fermentation of purple Jerusalem
artichoke extract to improve the α-glucosidase inhibitory effect in vitro and ame-
liorate blood glucose in db/db mice. Nutrition Research and Practice, 10(3), 282–287.
Yeo, J., & Shahidi, F. (2017). Effect of hydrothermal processing on changes of insoluble-
bound phenolics of lentils. Journal of Functional Foods, 38, 716–722.
Zhang, M. W., Zhang, R. F., Zhang, F. X., & Liu, R. H. (2010). Phenolic profiles and
antioxidant activity of black rice bran of different commercially available varieties.
Journal of Agricultural and Food Chemistry, 58, 7580–7587.
Zhang, X. Y., Chen, J., Li, X. L., Yi, K., Ye, Y., Liu, G., & Wang, Z. G. (2017). Dynamic
changes in antioxidant activity and biochemical composition of tartary buckwheat
leaves during Aspergillus niger fermentation. Journal of Functional Foods, 32, 375–381.