LWT - Food Science and Technology 122 (2020) 109064

Contents lists available at ScienceDirect

LWT - Food Science and Technology

journal homepage: www.elsevier.com/locate/lwt

Effects of lactic acid fermentation-based biotransformation on phenolic T

profiles, antioxidant capacity and flavor volatiles of apple juice
Caiyun Wua, Tianlin Lia, Jing Qia, Tian Jiangb, Huaide Xua, Hongjie Leia,∗
a
College of Food Science and Engineering, Northwest A&F University, Yangling, 712100, China
b
School of Food Science and Technology, Jiangnan University, Wuxi, 214122, China

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

Keywords: Apple juice (AJ) was fermented by six select lactic acid bacteria (LAB), namely Lactobacillus plantarum 90 (Lp90),
Apple juice Lactobacillus helveticus 76 (Lh76), Lactobacillus casei 37 (Lc37), Lactobacillus paracasei 01 (Lpc01), Lactobacillus
Probiotics acidophilus 85 (La85) and Bifidobacterium lactis 80 (Bla80). The effects of LAB fermentation on sugars and organic
Fermentation acids contents, phenolic profiles, antioxidant capacity and flavor volatiles of fermented AJ (FAJ) were in-
Antioxidant activity
vestigated. Results showed that AJ was an excellent food matrix for LAB strains. Lh76 and Lp90 showed the
Flavor volatiles
highest viable counts of 12.7 log CFU/mL at the end of fermentation. Lp90, La85 and Bla80 exhibited strong
malolactic conversion capacity and finally produced lactic acid more than 6.5 mg/mL. In addtion, LAB fer-
mentation significantly (p < 0.05) decreased the total phenols and flavonoids levels. However, antioxidant
activities based on DPPH and FRAP methods were enhanced during fermentation, especially for La85 which
exhibited dramatical DPPH radical scavenging activity (90.05%) and FRAP (0.96 mmol Trolox/L). The im-
provement of antioxidant activities was suggested to be related to the increase in caffeic acid and phlorizin
contents. The analysis of flavor volatiles showed that LAB fermentation resulted in formation of seven new
alcohols and six new esters, and improvement of ketones and aldehydes production.

1. Introduction confer the products with regulatory function of human intestinal mi-
croenvironment (Roberts et al., 2018; Xu et al., 2019). Probiotic fruit
China has become the world's largest apple producer after years of and vegetable products development is booming in recent years due to
development, with apple planting area and yield accounting for ap- the fact that the demand for functional foods has been dramatically
proximately 50% of the world. The production of apple in China was increased (Dimitrovski, Velickova, Langerholc, & Winkelhausen, 2015).
41.39 million tons in 2017, over 60% of which was Fuji apple cultivar Consumers, in particular vegans, lactose and casein intolerant in-
(FAO, 2017). The popularity of apple all over the world is due to not dividuals, and high cholesterol risk individuals, are increasingly inter-
only its pleasant flavor and established nutritional value, but also rich ested in consuming probiotic fruit and vegetable products (Di Cagno
in polyphenol compounds which are beneficial to human health et al., 2019; Li et al., 2019). Although some researches about the effects
(Laaksonen, Kuldjarv, Paalme, Virkki, & Yang, 2017). However, the of LAB fermentation on phytochemical and functional properties of
main consumption way of apple in China is commonly fresh fruit, plant based foods have been reported (Mousavi et al., 2013; Othman,
preserved fruit, concentrate juice and cider (Lorenzini, Simonato, Roblain, Chammen, Thonart, & Hamdi, 2009), there is a paucity of
Slaghenaufi, Ugliano, & Zapparoli, 2019; Ye, Yue, & Yuan, 2014). information about changes in flavor volatiles, phenolic profiles and
Therefore, it is necessary to explore advanced processing technologies antioxidant capacity of apple juice (AJ) after LAB fermentation.
for apple to enrich the product variety and enhance the health benefits. Based on the fact that different LAB strains behave strain specificity
Fruit and vegetable have been proposed as good media for the with different food matrix and exhibit different growth potential and
growth of lactic acid bacteria (LAB) because of the high contents of bioconversion capacity, the objective of this study was to investigate
vitamins, minerals, dietary fiber and antioxidant compounds the effects of six select commercial LAB strains fermentation on the
(Filannino, Di Cagno, & Gobbetti, 2018). LAB fermentation can improve physicochemical and phytochemical properties of AJ. Alterations in
the sensory and nutritional quality of fruit and vegetable juices, and growth capacity of LAB strains, sugars consumption, organic acids

∗
Corresponding author.
E-mail addresses: [email protected] (C. Wu), [email protected] (T. Li), [email protected] (J. Qi), [email protected] (T. Jiang),
[email protected] (H. Xu), [email protected] (H. Lei).

https://doi.org/10.1016/j.lwt.2020.109064
Received 10 November 2019; Received in revised form 15 January 2020; Accepted 16 January 2020
Available online 21 January 2020
0023-6438/ © 2020 Elsevier Ltd. All rights reserved.
C. Wu, et al. LWT - Food Science and Technology 122 (2020) 109064

production, phenolic profiles, antioxidant capacity, and flavor volatiles

during LAB fermentation were studied. Furthermore, the relationship
between phenolic compounds and antioxidant activities of FAJ was
evaluated.

2. Materials and methods

2.1. Microorganisms and culture conditions

Six commercial LAB strains namely Lactobacillus plantarum 90

(Lp90), Lactobacillus helveticus 76 (Lh76), Lactobacillus casei 37 (Lc37),
Lactobacillus paracasei 01 (Lpc01), Lactobacillus acidophilus 85 (La85)
and Bifidobacterium lactis 80 (Bla80) were kindly provided by WECARE-
BIO company (Jiangsu, China). These strains were maintained at
−80 °C in de Man, Rogosa and Sharpe (MRS) broth containing 30% of
glycerol (v/v). For activation, 1 mL of the culture was removed from
frozen storage and added to 10 mL of MRS broth, and incubated at 37 °C
for 16 h. The resulting LAB suspension was then used as inoculum for Fig. 1. Viable cell counts of select six lactic acid bacteria during fermentation of
100 mL of MRS broth and incubated at 37 °C for 16 h. Then the cells apple juice. Error bars indicate the standard deviations from two independent
were harvested by centrifugation (8000×g, 10 min, 4 °C), and washed samples. Abbreviations: Lp90, Lactobacillus plantarum 90; Lh76, Lactobacillus
twice with sterile water. The activated LAB cells were used for AJ fer- helveticus 76; Lc37, Lactobacillus casei 37; Lpc01, Lactobacillus paracasei 01;
mentation. La85, Lactobacillus acidophilus 85; Bla80, Bifidobacterium lactis 80.

2.2. Fermentation of apple juice

Fresh Fuji (Malus pumila Mill.) apples were purchased from a local
fruit market (Yangling, China). The apples were cleaned with fresh
water for several times and squeezed to get juice using a juicer (Midea,
China). Then AJ was obtained by filtration using a 100-mesh filter and
pH was adjusted to 5.0 using food-grade Na2CO3. The resulting AJ was
pasteurized at 72 °C for 15 min before fermentation. After cooling to
room temperature, 1% (v/v) of the activated LAB cells were allotted
into 2000 mL Erlenmeyer flasks containing 1000 mL of sterile AJ, and
incubated at 37 °C for 48 h. The sterile AJ with no inoculation treated
under the same conditions was used as control.

2.3. Determination of viable LAB counts

Viability tests were conducted by standard plate count method.

Serial dilutions (with 0.7% NaCl solution) of FAJ were prepared.
Aliquots of 0.1 mL of dilution were plated in MRS agar plates by spread
plate method. Then the plates were incubated at 37 °C for 48 h, and
plates containing 30–300 colonies were measured and recorded as log
CFU/mL.

2.4. Determination of soluble solids content (SSC), total sugars content, pH,
and titratable acidity (TA)

The SSC was measured using digital refractometer (Atago, Japan)

and expressed as deg;Brix. Total sugars content was analyzed as glucose
equivalents by the phenol sulfuric acid method. The pH was measured
using a pH meter (Mettler-Toledo, Switzerland). TA was determined by
titration with 0.01 mol/L NaOH and expressed as percentage of lactic
acid.

2.5. Antioxidant activities analysis

2.5.1. Determination of DPPH radical scavenging activity

DPPH radical scavenging was evaluated by previously reported
Fig. 2. Changes in total sugars (A) and soluble solids (B) during fermentation of
method (Li et al., 2019) with slight modifications. The sample (2 mL)
apple juice. Error bars indicate the standard deviations from two independent
was added to 4 mL of methanolic solution of DPPH (45 mg/L). Then,
samples. Values in the same pattern column with different superscript letters
the mixture was kept in the dark for 30 min and absorbance (A) was are significantly different (p < 0.05).
measured at 517 nm using UV-VIS spectrophotometer (Shimadzu,
Japan). Radical scavenging ability (RSA) was expressed as the percent
DPPH radical inhibition, and the formula was as follows:

2
C. Wu, et al. LWT - Food Science and Technology 122 (2020) 109064

Acontrol − Asample
DPPH RSA (%) = × 100
Acontrol

2.5.2. Determination of ABTS radical scavenging activity

ABTS radical scavenging was evaluated by previously described
method (Di Cagno et al., 2019) with slight modifications. Equivalent-
volume of ABTS solution (7 mmol/L) and K2S2O8 (2.45 mmol/L) was
mixture in the dark at room temperature for 16 h to produce ABTS
radical cation. Then the mixture was diluted with 80% ethanol to ob-
tain an absorbance of 0.70 ± 0.02 at 734 nm. Next, 400 μL of the
diluted sample (1:50) was added to 3.6 mL of the mixture and allowed
to react in the dark for 6 min and the absorbance was measured at
734 nm. RSA was expressed as the percent ABTS radical inhibition, and
the formula was as follows:
Acontrol − Asample
ABTS RSA (%) = × 100
Acontrol

2.5.3. Determination of FRAP

FRAP determination was according to the previously reported
method (Suárez-Jacobo et al., 2011) with slight modifications. The
reaction mixture was prepared by mixing 300 mL of acetate buffer
(0.3 mol/L, pH 3.6), 30 mL of 2,4,6-Tri(2-pyridyl)-1,3,5-triazine solu-
tion (10 mmol/L, prepared with 40 mmol/L HCl), and 30 mL of FeCl3
solution (20 mmol/L). The mixture was incubated in water bath at
37 °C for 30 min. Then, 0.1 mL of the diluted sample (1:10) was mixed
with 3 mL of the reaction mixture at 37 °C for 30 min, and the absor-
bance was measured at 593 nm. Results were expressed as mmol
Trolox/L using Trolox as a standard.

2.6. Phytochemical content analysis

2.6.1. Total phenolics content (TPC)

TPC was determined using the Folin-Ciocalteu method (Li et al.,
2019) with slight modifications. Briefly, 0.5 mL of the diluted sample
(1:16) was added to 2.5 mL of 10% Folin-Ciocalteu reagent. After re-
acting for 3 min, 2 mL of Na2CO3 (7.5% w/v) was added to the mixture.
Then put the mixture in dark for 60 min and the absorbance was
measured at 765 nm. Results were expressed as gallic acid equivalent
(GAE) (mg GAE/L).

2.6.2. Total flavonoids content (TFC)

TFC was determined using the AlCl3 colorimetric method (Kwaw
et al., 2018) with slight modifications. Briefly, 4 mL of the diluted
sample (1:5) was mixed with 0.5 mL of NaNO2 (50 g/L) for 5 min. After
that, 1 mL of AlCl3 (100 g/L) was added to the mixture and reacted for
5 min. Then 2 mL of NaOH (2 mol/L) was added to the mixture. The
absorbance was measured at 510 nm after 10 min. Results were ex-
pressed as rutin equivalent (RE) (mg RE/L).

2.7. Quantification of polyphenols and organic acids

Fig. 3. Changes in pH value (A), titratable acidity (B) and organic acids (C)
The HPLC system (Shimadzu, Japan) was used to quantify poly- during fermentation of apple juice. Error bars indicate the standard deviations
phenols and organic acids. The separation of compounds was carried from two independent samples. Values in the same pattern column with dif-
out on a C18 column (4.6 mm × 250 mm, 5 μm, Waters). Polyphenols ferent superscript letters are significantly different (p < 0.05).
were determined according to the previously reported method (Li et al.,
2019) with slight modifications. The mobile phase composed of 1% of
of 97:3 as mobile phase at 0.6 mL/min. UV–visible spectra detection
formic acid in water (solvent A) and acetonitrile (solvent B). The sol-
was set at 210 nm. The samples were filtered through 0.45 μm hydro-
vent gradient program was: 0–5 min, 5% B; 5–25 min, 12% B;
philic filter and the injection volume was 10 μL. Qualitative analysis of
25–40 min, 30% B; 40–50 min, 45% B; 50–60 min, 5% B. Flow rate was
polyphenols and organic acids was carried out by comparing the peak
1 mL/min. Column oven temperature was 30 °C. UV–visible spectra
areas to the linear calibration curves of the external standards (Sigma-
detection was set at 280 nm. Organic acids were determined using the
Aldrich, USA).
previously reported method (Ye et al., 2014) with slight modifications.
Elution was conducted in the isocratic mode by using the mixture of
0.01 mol/L KH2PO4–H3PO4 (pH 2.7) and methyl alcohol with the ratio

3
C. Wu, et al. LWT - Food Science and Technology 122 (2020) 109064

2.9. Determination of volatile compounds

Volatile compounds were analyzed by HS-SPME-GC-MS according

to the previously reported method (Kaprasob, Kerdchoechuen,
Laohakunjit, Sarkar, & Shetty, 2017) with slight modifications. Sample
(5 mL) was added into 15 mL glass vial containing 2 g of NaCl, then
10 μL of 2-Octanol (Sigma-Aldrich, USA) was added as internal stan-
dard (0.2 mg/mL, prepared with ethanol). The headspace micro-
extraction was performed with a 75 μm Carboxen-polydimethylsiloxane
(CAR-PDMS) fiber. The samples were performed for 30 min at 40 °C
after 15 min equilibration. The analysis was performed by GC-MS
system (Thermo Finnigan, USA) equipped with a TR-5MS column
(30 m × 0.25 mm, 0.25 μm, J&W Scientific, USA) and a quadrupole
DSQ II MS. The oven temperature program was 3 min at 50 °C and then
5 °C/min to 200 °C for 12 min. Helium was used as carrier gas with a
flow rate of 1 mL/min. The semi-quantification of volatile compounds
was performed on the basis of internal standard (2-Octanol).

2.10. Statistical analysis

Statistical calculation was carried out by SPSS 18 (SPSS Inc., USA)

for one-way ANOVA. Student-Newman-Keuls test was applied for the
comparison of mean values and identification of significant differences
(p < 0.05) among treatments.

3. Results and discussion

3.1. Physicochemical properties during fermentation

LAB viable counts in AJ during LAB fermentation are shown in

Fig. 1. Six LAB strains showed the similar growth profile within 24 h
and reached the maximum values at 36 h of fermentation. At the end of
fermentation, Lh76 and Lp90 showed the maximum viable counts of
12.74 log CFU/mL. An obvious decrease in the microbial population
was observed with La85 and Bla80 after 36 h of fermentation, due to
the fact that lower pH environment created by LAB fermentation has a
Fig. 4. Changes in total phenols content (A) and total flavonoids content (B) stress impact on the growth of LAB strains (Kaprasob et al., 2017). This
during fermentation of apple juice. Error bars indicate the standard deviations result was consistent with previous study in which strong and inverse
from two independent samples. Values in the same pattern column with dif- correlation between pH value and LAB viable counts was observed (Li
ferent superscript letters are significantly different (p < 0.05). et al., 2019).
As shown in Fig. 2A, the contents of total sugars in AJ were sig-
2.8. Color assessment nificantly (p < 0.05) decreased by 17–23% during LAB fermentation,
which can be attributed to its utilization by LAB strains for cellular
The CIE color parameters (L*, a* and b*) were measured using the growth and bioconversion into lactic acid (Verón et al., 2019). The SSC
Hunter colorimeter model color Quest XE (HunterLab, USA). The col- may be considered as a balance between the consumption of sugars and
orimeter was calibrated with white and black standard tiles. Then the the production of organic acids during fermentation (Maldonado et al.,
sample was placed in a transparent cuvette for the detection to obtain 2017). There was an obvious decrease in SSC in the first 24 h of fer-
L*, a*, b* values. Total color difference (ΔE) was calculated by the mentation (Fig. 2B), indicating the rapid consumption of sugars by LAB
following formula: strains (Kwaw et al., 2018), then a slight decrease was observed during
24–48 h of fermentation, and the final SSC of FAJ was about
ΔE = (L0 − L)2 + (a0 − a)2 + (b0 − b)2 12.1deg;Brix.
As expected, the pH value was decreased from 5.26 to 3.91 during

Table 1
Phenolic profiles of apple juice and lactic-acid-fermented apple juice.
Control Lp90 Lh76 Lc37 Lpc01 La85 Bla80

Gallic acid 1.61 ± 0.011d 1.83 ± 0.018b 3.17 ± 0.018a 1.82 ± 0.021b 1.79 ± 0.011bc 1.78 ± 0.008c 1.80 ± 0.012bc
Chlorogenic acid 3.92 ± 0.064d 3.60 ± 0.023e 3.51 ± 0.017f 4.87 ± 0.018a 4.54 ± 0.009b 4.44 ± 0.012c 4.62 ± 0.013b
Caffeic acid 1.17 ± 0.008d 1.22 ± 0.009b 1.21 ± 0.010b 1.26 ± 0.008a 1.15 ± 0.013d 1.18 ± 0.007cd 1.20 ± 0.009bc
Epicatechin 3.04 ± 0.063f 3.27 ± 0.012d 4.20 ± 0.016a 3.46 ± 0.013bc 3.52 ± 0.017b 3.19 ± 0.007e 3.44 ± 0.016c
p-Coumaric acid 1.34 ± 0.011ab 1.36 ± 0.007ab 1.28 ± 0.009c 1.27 ± 0.011c 1.35 ± 0.011ab 1.37 ± 0.019a 1.34 ± 0.012b
Ferulic acid 0.70 ± 0.016a 0.67 ± 0.007b 0.61 ± 0.013d 0.66 ± 0.006bc 0.62 ± 0.009d 0.66 ± 0.005bc 0.64 ± 0.010cd
Ellagic acid 0.89 ± 0.011a 0.85 ± 0.013b 0.84 ± 0.004b 0.85 ± 0.006b 0.86 ± 0.011ab 0.86 ± 0.011ab 0.85 ± 0.015b
Rutin 2.46 ± 0.016b 2.48 ± 0.071b 2.45 ± 0.016b 2.47 ± 0.013b 2.62 ± 0.020a 2.62 ± 0.005a 2.08 ± 0.010c
Phlorizin 5.54 ± 0.012d 6.17 ± 0.009b 6.19 ± 0.008b 6.18 ± 0.025b 6.18 ± 0.007b 5.79 ± 0.008c 6.47 ± 0.013a

4
C. Wu, et al. LWT - Food Science and Technology 122 (2020) 109064

The changes in two main organic acids contents (malic acid and
lactic acid) during LAB fermentation are shown in Fig. 3C. Six LAB
strains exhibited dramatic malolactic conversion capacity, and malic
acid was depleted by LAB strains at the end of fermentation. La85, Lp90
and Bla80 produced the highest content of lactic acid more than 6 g/L,
as compared with Lc37 and Lpc01 showing the lowest lactic acid pro-
duction (ca. 3 g/L). The content of lactic acid was much higher than
malic acid because of another biosynthesis pathway of lactic acid from
sugars by LAB metabolism (Cousin et al., 2017).

3.2. Phytochemical concentration during fermentation

The TPC and TFC in fruit and vegetable are important character-
istics due to their positive impact on various physiological processes
related to human health, such as cancer risk reduction, anti-in-
flammatory properties, and inhibition of carcinogenesis (de Souza, de
Albuquerque, dos Santos, Massa, & de Brito Alves, 2019; Li et al., 2017).
It has been reported that TPC and TFC in fruit juice were increased
during fermentation with Lactobacillus plantarum MCC 2974 (Vivek,
Mishra, Pradhan, & Jayabalan, 2019). However, in this study, there was
a dramatic decrease in TPC and TFC in AJ during LAB fermentation,
especially for Lh76 decreasing TPC and TFC by 26.9% and 33.2%, re-
spectively (Fig. 4A and 4B). This result can be attributed to the depo-
lymerization of macromolecular phenolic or conversion of individual
phenolic compounds conducted by LAB strains (Kaprasob et al., 2017;
Li et al., 2019). In addition, it has been proposed that the increase of
free form of phenolic compounds and the production of other by-
products during LAB fermentation might be responsible for the im-
provement of antioxidant capacity (Kwaw et al., 2018). To confirm this
hypothesis, the phenolic profiles and antioxidant activities of FAJ were
further investigated in this study.

3.3. Phenolic profiles and antioxidant activities

As shown in Table 1, phlorizin and chlorogenic acid were the

dominant phenolic compounds in AJ, with contents of 5.54 and
3.92 mg/L, respectively. LAB fermentation significantly (p < 0.05)
increased the contents of gallic acid, epicatechin and phlorizin, while
decreased the contents of ferulic acid and ellagic acid, especially for
Lh76 fermentation which increased the contents of gallic acid and
epicatechin by 87% and 38%, respectively, and Bla80 fermentation
increased phlorizin content by 17%. Despite several phenolic com-
pound molecules with similar skeleton or structural characteristics, the
number and position of hydroxyl groups play a key role in biological
and functional properties (de Souza et al., 2019). It has been confirmed
that phenolic compounds, as reducing agents, free-radical scavengers,
and singlet oxygen quenchers, possessed remarkable antioxidant ac-
tivity mainly attributing to the hydrogen atom transfer or electron
donation to free radicals (Verón et al., 2019). Therefor, antioxidant
activities based on DPPH, ABTS and FRAP methods were analyzed in
this study. Although LAB fermentation decreased ABTS radical
scavenging activity, it significantly increased the DPPH radical
scavenging activity and FRAP of FAJ (Fig. 5A, B and C). Lp90, La85,
Fig. 5. Antioxidant activities based on DPPH (A), ABTS (B) and FRAP (C)
and Bla80 fermentation exhibited higher DPPH radical scavenging ac-
methods during fermentation of apple juice. Error bars indicate the standard
tivity, increasing by 6.4%, 5.2%, and 6.3%, respectively, as compared
deviations from two independent samples. Values in the same pattern column
with different superscript letters are significantly different (p < 0.05). with control. In the case of Lh76 and Lc37 fermentation, there were
approximately 1% and 3% decreases in DPPH radical scavenging ac-
tivity after 12 h of fermentation, respectively, then reached the max-
fermentation, indicating that more acid compounds were produced by
imum after 36 h of fermentation. Only Lpc01 fermentation showed no
LAB strains (Fig. 3A). The variation in TA was negatively correlated
significant change in DPPH radical scavenging activity (p > 0.05). The
with the change in pH value (Chen, Lu, Yu, Chen, & Tian, 2019). The TA
increase in DPPH radical scavenging activity suggested that LAB fer-
of all samples showed a slight decrease within 12 h (Fig. 3B) due to the
mentation might improve the availability of polyphenol compounds
conversion of malic acid into lactic acid at initial stage of LAB fer-
with proton-donating properties (dos Santos Filho et al., 2019). Fur-
mentation (Markkinen, Laaksonen, Nahku, Kuldjarv, & Yang, 2019).
thermore, significantly positive correlations between the contents of
Compared with other strains, Lc37 and Lh76 fermentation increased TA
caffeic acid and phlorizin and DPPH radical scavenging activity were
significantly (p < 0.05) at the end of fermentation (ca. 0.5 g/100 mL).
found (p < 0.05) (Table 2). It has been confirmed that LAB can convert

5
C. Wu, et al. LWT - Food Science and Technology 122 (2020) 109064

Table 2
Pearson's correlation coefficients of phenolic profiles and antioxidant activities.
Gallic acid Chlorogenic acid Caffeic acid Epicatechin p-Coumaric acid Ferulic acid Ellagic acid Rutin Phlorizin DPPH ABTS FRAP

Gallic acid 1
Chlorogenic acid −0.538* 1
Caffeic acid 0.237 0.066 1
Epicatechin 0.924** −0.273 0.265 1
p-Coumaric acid −0.538* −0.072 −0.627** −0.635** 1
Ferulic acid −0.579** −0.091 −0.029 −0.791** 0.353 1
Ellagic acid −0.451* −0.103 −0.363 −0.590** 0.436* 0.697** 1
Rutin −0.032 −0.098 −0.247 −0.108 0.161 0.028 0.217 1
Phlorizin 0.273 0.220 0.401 0.535* −0.345 −0.700** −0.733** −0.527* 1
DPPH 0.150 0.047 0.677** 0.153 −0.141 −0.135 −0.545* −0.385 0.532* 1
ABTS −0.359 0.163 −0.389 −0.465* 0.055 0.548* 0.504* −0.102 −0.615** −0.479* 1
FRAP 0.016 0.141 0.272 −0.113 0.188 0.166 −0.118 −0.054 −0.073 0.556** −0.025 1

*Correlation is significant at p < 0.05. **Correlation is significant at p < 0.01.Correlation is significant at p < 0.05. **Correlation is significant at p < 0.01.

Table 3 contribution of each polyphenolic compound to antioxidant activities

Colorimetric properties of apple juice and lactic-acid-fermented apple juice. (Kwaw et al., 2018).
L* a* b* ΔE
3.4. Color properties
Control 81.20 ± 0.05e
3.09 ± 0.26a
32.29 ± 6.73 a
–
Lp90 87.70 ± 0.10a 0.75 ± 0.02cd 25.14 ± 0.07c 4.53 ± 0.09b
Lh76 86.40 ± 0.03c 0.83 ± 0.00c 25.59 ± 0.04c 3.44 ± 0.01d The effect of LAB fermentation on color properties of FAJ was stu-
Lc37 88.00 ± 0.14a 0.33 ± 0.02e 24.62 ± 0.06d 4.91 ± 0.14a died by the following color features (L*, a* and b*) and total color
Lpc01 82.54 ± 0.17d 1.62 ± 0.02b 26.45 ± 0.03b 2.30 ± 0.09e difference (ΔE). Results showed that a* and b* were decreased and L*
La85 87.11 ± 0.01b 0.43 ± 0.01de 24.39 ± 0.03d 4.06 ± 0.02c was increased after LAB fermentation (Table 3). The color of probiotic
Bla80 87.97 ± 0.07a 0.41 ± 0.03de 24.50 ± 0.09d 4.85 ± 0.08a
fruit and vegetable products can be affected by phenolic contents and
Results are the mean value of two replicates. L*–Lightness, a*–Redness, profiles (Cousin et al., 2017; Laaksonen et al., 2017; Wang et al., 2015).
b*–Yellowness, ΔE–color difference. Values in the same column with different In this study, the decreases in a* and b* were related to the changes in
superscript letters are significantly different (p < 0.05). phenolic compounds of FAJ. Results of correlation analysis showed that
TPC and TFC were significantly and positively correlated with a* and
Table 4 b*, and negatively correlated with L* (p < 0.01) (Table 4). In addition,
Pearson's correlation coefficients of phytochemical concentration and color there were significant differences in ΔE values between AJ and FAJ
attributes. (p < 0.05) (Table 3). LAB fermentation showed dramatical changes in
TPC TFC L* a* b*
ΔE values, except for Lpc01 fermentation with the smallest ΔE, sug-
gesting the color closest to AJ.
TPC 1
TFC 0.808** 1
L* −0.574** −0.753** 1
3.5. Flavor volatiles
a* 0.699** 0.910** −0.936** 1
b* 0.799** 0.975** −0.817** 0.944** 1 One of the most important characteristics of fruit juice for consumer
palatability is the flavor volatile compounds (Kaprasob et al., 2017). In
*Correlation is significant at p < 0.05. **Correlation is significant at this study, a total of 52 kinds of volatile compounds were identified in
p < 0.01.
AJ and FAJ (Table 5). Results showed that alcohols and esters were
quantitatively the major flavor volatile compounds. Alcohols were im-
caffeic acid to dihydrocaffeic acid by phenolic acid reductase, an in- portant volatile aroma in the process of LAB fermentation. A small
tracellular antioxidant with more potent antioxidant than caffeic acid amount of alcohols impart light aroma and could be used as a good
(Filannino et al., 2018; Ricci et al., 2019). In addition, significantly solvent for other aroma substances (Laaksonen et al., 2017). In this
negative correlation between DPPH radical scavenging activity and study, 13 kinds of alcohols were identified and the main alcohol com-
ellagic acid content was observed (p < 0.05). pounds were hexyl alcohol, ethanol, 2-methyl-1-butanol, and seven
As show in Fig. 5B, ABTS radical scavenging activity showed a de- new alcohols appeared after LAB fermentation, including 2-methyl-1-
crease trend during LAB fermentation, which was inconsistent with the propanol, isobutenylcarbinol, trans-2-hexen-1-ol, benzyl alcohol, 1-oc-
results of DPPH radical scavenging activity. This can be attributed to tanol, β-citronellol and geraniol. It has been reported that a great
the different antioxidant mechanism that ABTS is based on hydrogen number of 3-methyl-1-butanl could cause uncomfortable taste of FAJ
atom transfer mechanism, whereas DPPH is based on electron transfer (Cousin et al., 2017). However, LAB fermentation significantly reduced
mechanism (Kaprasob et al., 2017). The decrease in ABTS radical the 3-methyl-1-butanl content, especially for La85 fermentation, which
scavenging activity is related to the oxidation and degradation of an- decreased 3-methyl-1-butanl content by 93%. In addition, the new al-
tioxidant or the synergistic and redox interactions among different cohols appeared after fermentation enriched the background flavors of
compounds (Alam, Bristi, & Rafiquzzaman, 2013). Correlation results FAJ, especially for β-citronellol appeared after Lc37, Lpc01 and Bla80
showed that ABTS radical scavenging activity was significantly and fermentation, which was a terpene alcohol compound with rose and
positively correlated with the contents of ferulic acid and ellagic acid, citrus floral notes.
while negatively correlated with the contents of epicatechin and Esters have a significant contribution to the aroma of FAJ. In this
phlorizin (p < 0.05) (Table 2). In addition, LAB fermentation posi- study, 24 kinds of esters were detected, and the highest content was 2-
tively affected the FRAP of FAJ (Fig. 5C). La85 fermentation sig- methylbutyl acetate, followed by butyl acetate, ethyl butyrate, ethyl 2-
nificantly (p < 0.05) increased FRAP by 9.4%, as compared with methylbutyrate and hexyl acetate. Although the total amount of esters
control. The differences in antioxidant activities could be the results of was decreased after LAB fermentation, six new esters were observed,
the impact of LAB fermentation on phenolic profiles and the including ethyl acetate, ethyl propionate, prenyl acetate, isobutyl

6
C. Wu, et al. LWT - Food Science and Technology 122 (2020) 109064

Table 5
Concentration of flavor volatiles in apple juice and lactic-acid-fermented apple juice (μg/L).
Flavor volatiles Control Lp90 Lh76 Lc37 Lpc01 La85 Bla80

Alcohols
1 Ethanol 188.37 109.94 123.48 111.58 101.00 91.98 83.42
2 2-Methyl-1-propanol – – – 8.17 – – –
3 1-Butanol 42.55 44.90 49.00 31.25 27.61 30.00 25.89
4 Isobutenylcarbinol – 1.58 1.10 – 0.80 – 0.65
5 3-Methyl-1-butanol 73.18 5.04 5.04 16.90 16.45 4.47 14.67
6 2-Methyl-1-butanol 128.86 103.02 113.55 74.69 66.11 72.68 64.97
7 1-Pentanol 3.07 5.28 5.52 3.48 3.01 3.42 3.15
8 trans-2-Hexen-1-ol – 65.02 125.38 – – – –
9 Hexyl alcohol 313.48 329.90 346.63 341.44 189.30 312.51 320.27
10 Benzyl alcohol – 2.34 3.07 1.67 – 2.50 2.66
11 1-Octanol – – – – 1.24 2.35 2.15
12 β-Citronellol – – – 2.61 1.46 – 4.64
13 Geraniol – 1.87 1.70 – – – –
Subtotal 749.51 668.90 774.48 591.79 406.98 519.90 522.46
Esters
1 Ethyl acetate – – 111.84 82.71 26.86 49.06 44.73
2 Ethyl propionate – 25.30 28.58 16.72 16.08 15.25 –
3 Propyl acetate 45.14 35.84 37.86 23.63 24.21 22.68 19.38
4 Methyl butyrate 4.08 4.76 4.79 2.76 3.01 2.76 2.24
5 Isobutyl acetate 18.74 15.03 15.98 9.33 9.22 8.58 7.73
6 Methyl 2-methylbutyrate 7.09 4.65 5.05 3.17 3.24 2.88 2.43
7 Ethyl butyrate 152.57 111.57 118.00 71.99 71.44 68.55 62.21
8 Propyl propionate 16.87 12.75 13.81 – 7.59 7.27 6.88
9 Butyl acetate 240.10 194.97 211.61 131.72 127.81 123.68 111.81
10 Ethyl 2-methylbutyrate 140.93 82.80 85.69 53.10 52.22 47.93 43.27
11 2-Methylbutyl acetate 532.14 423.83 463.69 284.45 255.69 255.04 240.08
12 Propyl butyrate 31.78 27.06 28.49 16.16 15.03 13.97 15.86
13 Butyl propionate 6.18 4.85 5.24 – 3.29 – –
14 Amyl acetate 38.84 31.74 33.79 19.00 17.30 17.42 17.66
15 Prenyl acetate – – – 1.30 – – 1.02
16 Methyl hexanoate 0.29 0.49 0.59 0.36 0.19 – 0.65
17 Propyl 2-methylbutanoate 47.55 23.12 24.65 14.21 12.73 12.32 11.85
18 Isobutyl butyrate – 0.35 – – – – –
19 Ethyl caproate 16.80 9.61 9.24 5.20 4.33 5.10 5.47
20 Hexyl acetate 101.76 71.30 74.42 45.85 42.33 46.87 46.85
21 Butyl 2-methylbutanoate 8.57 3.27 3.29 1.85 1.64 1.76 1.69
22 Ethyl caprylate 1.31 4.95 2.43 3.21 1.71 0.99 1.11
23 Phenethyl acetate – 0.89 0.62 0.31 – 0.20 0.23
24 Ethyl caprate – 0.85 – 0.49 – – –
Subtotal 1410.74 1089.96 1279.63 787.52 695.92 702.33 643.13
Aldehydes
1 Acetaldehyde – 5.35 2.11 4.53 2.36 1.88 3.00
2 Hexanal 2.46 4.09 3.45 1.37 0.80 2.13 1.84
3 (E)-2-Hexenal – 7.01 6.42 – – – –
4 Nonanal – 4.85 – – – – –
5 Decanal – 1.08 1.36 1.68 1.22 1.40 1.13
Subtotal 2.46 22.38 13.33 7.57 4.38 5.41 5.96
Ketones
1 2,3-Butanedione – – 0.86 – – 1.25 2.31
2 2-Pentanone – – 0.60 1.30 1.32 0.96 0.53
3 2,3-Pentanedione – – – 1.80 1.73 – 1.50
4 Acetoin 5.43 – 1.15 – 4.26 0.83 –
5 2-Heptanone – 3.98 11.89 11.98 16.90 11.40 7.03
6 Methylheptenone – 3.75 3.07 2.37 2.64 2.23 2.52
7 2-Octanone 9.18 4.81 3.91 6.03 8.48 – 7.20
8 2-Nonanone 1.28 1.15 2.40 3.14 18.64 2.26 1.16
9 2-Undecanone 0.68 – – 0.51 0.22 – 0.32
10 Damascenone – 3.32 3.75 2.44 2.42 3.56 3.88
Subtotal 16.57 17.02 27.62 29.57 56.61 22.48 26.44

Results are the mean value of two replicates. Standard errors were always lower than 10% of mean value. −, Not detectable.

butyrate, phenethyl acetate and ethyl caprate, which increased the 4. Conclusions
aroma complexity of FAJ. In addition, ten ketones and five aldehydes
were detected in AJ and FAJ, and four new aldehydes and six new AJ, an excellent food matrix for LAB strains, was explored for the
ketones appeared after LAB fermentation. The (E)-2-hexenal appeared development of a possible probiotic product with large market potential
after Lp90 and Lh76 fermentation exhibited a pleasant green leaf fra- in this study. AJ was fermented by six select commercial LAB strains.
grance. Methylheptenone appeared after LAB fermentation showed the All the strains exhibited good growth capacity with 11 log CFU/mL in
aroma of lemon grass. Therefore, FAJ possessed a strong floral scent FAJ, higher than 8 log CFU/mL, a cell load that has been proven to
and fragrance, and LAB fermentation was suggested to be an effective possess health benefits. La85 fermentation exhibited dramatical mal-
way to enrich the aroma of FAJ. olactic conversion and possessed strong antioxidant activities based on
DPPH and FRAP methods. In addition, DPPH radical scavenging

7
C. Wu, et al. LWT - Food Science and Technology 122 (2020) 109064

activity was positively correlated with the contents of caffeic acid and Fermentation-based biotransformation of bioactive phenolics and volatile compounds
phlorizin, while negatively correlated with ellagic acid content. Seven from cashew apple juice by select lactic acid bacteria. Process Biochemistry, 59,
141–149.
new alcohols, six new esters, four new aldehydes and six new ketones Kwaw, E., Ma, Y., Tchabo, W., Apaliya, M. T., Wu, M., Sackey, A. S., et al. (2018). Effect of
compounds appeared after LAB fermentation, suggesting the improve- lactobacillus strains on phenolic profile, color attributes and antioxidant activities of
ment of aroma complexity in FAJ. Our further study should focus on lactic-acid-fermented mulberry juice. Food Chemistry, 250, 148–154.
Laaksonen, O., Kuldjarv, R., Paalme, T., Virkki, M., & Yang, B. (2017). Impact of apple
assessing the storage properties, and gastric and intestinal digestibility cultivar, ripening stage, fermentation type and yeast strain on phenolic composition
of FAJ, to provide more valuable information about the physiological of apple ciders. Food Chemistry, 233, 29–37.
functions of FAJ to human health. Li, Z., Teng, J., Lyu, Y., Hu, X., Zhao, Y., & Wang, M. (2019). Enhanced antioxidant
activity for apple juice fermented with Lactobacillus plantarum ATCC14917.
Molecules, 24, 51.
CRediT authorship contribution statement Li, X., Xing, Y., Cao, L., Xu, Q., Li, S., Wang, R., et al. (2017). Effects of six commercial
Saccharomyces cerevisiae strains on phenolic attributes, antioxidant activity, and
aroma of kiwifruit (Actinidia deliciosa cv.) wine. BioMed Research International,
Caiyun Wu: Conceptualization, Methodology, Software,
2934743 2017.
Investigation, Writing - original draft. Tianlin Li: Validation, Formal Lorenzini, M., Simonato, B., Slaghenaufi, D., Ugliano, M., & Zapparoli, G. (2019).
analysis, Data curation. Jing Qi: Validation, Formal analysis, Assessment of yeasts for apple juice fermentation and production of cider volatile
Visualization. Tian Jiang: Resources, Writing - review & editing. compounds. LWT-Food Science and Technology, 99, 224–230.
Maldonado, R. R., Araújo, L.d. C., Dariva, L. C.d. S., Rebac, K. N., Pinto, I. A.d. S., Prado,
Huaide Xu: Resources, Writing - review & editing, Supervision. J. P. R., et al. (2017). Potential application of four types of tropical fruits in lactic
Hongjie Lei: Validation, Resources, Data curation, Writing - review & fermentation. LWT-Food Science and Technology, 86, 254–260.
editing, Supervision, Funding acquisition. Markkinen, N., Laaksonen, O., Nahku, R., Kuldjarv, R., & Yang, B. (2019). Impact of lactic
acid fermentation on acids, sugars, and phenolic compounds in black chokeberry and
sea buckthorn juices. Food Chemistry, 286, 204–215.
Declaration of competing interest Mousavi, Z. E., Mousavi, S. M., Razavi, S. H., Hadinejad, M., Emam-Djomeh, Z., &
Mirzapour, M. (2013). Effect of fermentation of pomegranate juice by Lactobacillus
plantarum and Lactobacillus acidophilus on the antioxidant activity and metabolism
None. of sugars, organic acids and phenolic compounds. Food Biotechnology, 27, 1–13.
Results are the mean value of two replicates. Values in the same row Othman, N. B., Roblain, D., Chammen, N., Thonart, P., & Hamdi, M. (2009). Antioxidant
with different superscript letters are significantly different (p < 0.05). phenolic compounds loss during the fermentation of Chétoui olives. Food Chemistry,
116, 662–669.
Abbreviations: Lp90, Lactobacillus plantarum 90; Lh76, Lactobacillus Ricci, A., Cirlini, M., Calani, L., Bernini, V., Neviani, E., Del Rio, D., et al. (2019). In vitro
helveticus 76; Lc37, Lactobacillus casei 37; Lpc01, Lactobacillus paracasei metabolism of elderberry juice polyphenols by lactic acid bacteria. Food Chemistry,
01; La85, Lactobacillus acidophilus 85; Bla80, Bifidobacterium lactis 80. 276, 692–699.
Roberts, D., Reyes, V., Bonilla, F., Dzandu, B., Liu, C., Chouljenko, A., et al. (2018).
Viability of Lactobacillus plantarum NCIMB 8826 in fermented apple juice under
Acknowledgments simulated gastric and intestinal conditions. LWT-Food Science and Technology, 97,
144–150.
The authors gratefully acknowledge the National Natural Science dos Santos Filho, A. L., Freitas, H. V., Rodrigues, S., Abreu, V. K. G., de Oliveira Lemos, T.,
Gomes, W. F., et al. (2019). Production and stability of probiotic cocoa juice with
Foundation of China (No. 31501467) for the financial support. sucralose as sugar substitute during refrigerated storage. LWT-Food Science and
Technology, 99, 371–378.
References de Souza, E. L., de Albuquerque, T. M. R., dos Santos, A. S., Massa, N. M. L., & de Brito
Alves, J. L. (2019). Potential interactions among phenolic compounds and probiotics
for mutual boosting of their health-promoting properties and food functionalities - a
Alam, M. N., Bristi, N. J., & Rafiquzzaman, M. (2013). Review on in vivo and in vitro review. Critical Reviews in Food Science and Nutrition, 59, 1645–1659.
methods evaluation of antioxidant activity. Saudi Pharmaceutical Journal, 21, Suárez-Jacobo, Á., Rüfer, C. E., Gervilla, R., Guamis, B., Roig-Sagués, A. X., & Saldo, J.
143–152. (2011). Influence of ultra-high pressure homogenisation on antioxidant capacity,
Chen, C., Lu, Y., Yu, H., Chen, Z., & Tian, H. (2019). Influence of 4 lactic acid bacteria on polyphenol and vitamin content of clear apple juice. Food Chemistry, 127, 447–454.
the flavor profile of fermented apple juice. Food Bioscience, 27, 30–36. Verón, H. E., Cano, P. G., Fabersani, E., Sanz, Y., Isla, M. I., Fernández Espinar, M. T.,
Cousin, F. J., Le Guellec, R., Schlusselhuber, M., Dalmasso, M., Laplace, J. M., & Cretenet, et al. (2019). Cactus pear (Opuntia ficus-indica) juice fermented with autochthonous
M. (2017). Microorganisms in fermented apple beverages: Current knowledge and Lactobacillus plantarum S-811. Food & Function, 10, 1085–1097.
future directions. Microorganisms, 5, 39. Vivek, K., Mishra, S., Pradhan, R. C., & Jayabalan, R. (2019). Effect of probiotification
Di Cagno, R., Filannino, P., Vincentini, O., Cantatore, V., Cavoski, I., & Gobbetti, M. with Lactobacillus plantarum MCC 2974 on quality of Sohiong juice. LWT-Food
(2019). Fermented portulaca oleracea L. Juice: A novel functional beverage with Science and Technology, 108, 55–60.
potential ameliorating effects on the intestinal inflammation and epithelial injury. Wang, L., Sun, X., Li, F., Yu, D., Liu, X., Huang, W., et al. (2015). Dynamic changes in
Nutrients, 11, 248. phenolic compounds, colour and antioxidant activity of mulberry wine during alco-
Dimitrovski, D., Velickova, E., Langerholc, T., & Winkelhausen, E. (2015). Apple juice as a holic fermentation. Journal of Functional Foods, 18, 254–265.
medium for fermentation by the probiotic Lactobacillus plantarum PCS 26 strain. Xu, X., Bao, Y., Wu, B., Lao, F., Hu, X., & Wu, J. (2019). Chemical analysis and flavor
Annals of Microbiology, 65, 2161–2170. properties of blended orange, carrot, apple and Chinese jujube juice fermented by
FAO (2017). FaoStat database. http://faostat.fao.org. selenium-enriched probiotics. Food Chemistry, 289, 250–258.
Filannino, P., Di Cagno, R., & Gobbetti, M. (2018). Metabolic and functional paths of Ye, M., Yue, T., & Yuan, Y. (2014). Evolution of polyphenols and organic acids during the
lactic acid bacteria in plant foods: Get out of the labyrinth. Current Opinion in fermentation of apple cider. Journal of the Science of Food and Agriculture, 94,
Biotechnology, 49, 64–72. 2951–2957.
Kaprasob, R., Kerdchoechuen, O., Laohakunjit, N., Sarkar, D., & Shetty, K. (2017).