Received: 5 November 2020 Accepted: 15 November 2020

DOI: 10.1002/tox.23065

RESEARCH ARTICLE

Extracts of Jasminum sambac flowers fermented by

Lactobacillus rhamnosus inhibit H2O2- and UVB-induced aging
in human dermal fibroblasts

Chih-Chu Ho1 | Shang-Chuan Ng1 | Ho-Lin Chuang1 | Su-Ying Wen2,3,4 |

5 6 7,8,9,10,11
Chia-Hua Kuo | B. Mahalakshmi | Chih-Yang Huang | Wei-Wen Kuo1

1
Department of Biological Science and
Technology, College of Biopharmaceutical and Abstract
Food Sciences, China Medical University, Ultraviolet (UV) irradiation is a crucial factor that leads to skin photoaging and results
Taichung, Taiwan
2 in increased DNA damage, oxidative stress, and collagen degradation. Jasmine
Department of Dermatology, Taipei City
Hospital, Renai Branch, Taipei, Taiwan flowers have been utilized as a traditional medicine in Asia to treat various diseases,
3
Department of Cosmetic Applications and including dermatitis, diarrhea, and fever. Furthermore, the fermented broth of Lacto-
Management, Mackay Junior College of
Medicine, Nursing, and Management, Taipei, bacillus rhamnosus has been reported to exert protective effects on the skin. In the
Taiwan present study, jasmine flower extract was fermented with L. rhamnosus. We investi-
4
Department of Health Care Management,
gated the antioxidant and collagen-promoting effects on UVB/H2O2-induced HS68
National Taipei University of Nursing and
Health Sciences, Taipei City, Taiwan dermal fibroblast cell damage. The results indicated that treatment with the fer-
5
Department of Sports Sciences, University of mented flower extracts of Jasminum sambac (F-FEJS) could enhance the viability of
Taipei, Taipei, Taiwan
6
HS68 cells. Furthermore, the UVB/H2O2-induced excessive production of reactive
Institute of Research and Development, Duy
Tan University, Da Nang, Vietnam oxygen species, degradation of collagen, activation of MAPKs, including P38, ERK,
7
Graduate Institute of Biomedical Sciences, and JNK, and premature senescence were remarkably attenuated by F-FEJS in der-
China Medical University, Taichung, Taiwan
mal fibroblast cells. The nuclear accumulation of p-c-jun, which is downstream of
8
Cardiovascular and Mitochondrial Related
Disease Research Center, Hualien Tzu Chi MAPK, and the inactivation of p-smad2/3, which is one of the crucial transcription
Hospital, Buddhist Tzu Chi Medical factors that enhance collagen synthesis, were reversed in response to F-FEJS treat-
Foundation, Hualien, Taiwan
9
ment in UVB/H2O2-exposed cells. Notably, the expression of antioxidant genes, such
Center of General Education, Buddhist Tzu
Chi Medical Foundation, Tzu Chi University of as HO-1, and the nuclear translocation of Nrf2 were further enhanced by F-FEJS in
Science and Technology, Hualien, Taiwan
UVB/H2O2-treated cells. Interestingly, the F-FEJS-induced increase in ARE luciferase
10
Department of Medical Research, China
Medical University Hospital, China Medical
activity indicated the activation of Nrf2/ARE signaling. In conclusion, our findings
University, Taichung, Taiwan demonstrated that F-FEJS can effectively ameliorate UVB/H2O2-induced dermal cell
11
Department of Biotechnology, Asia aging and may be considered a promising ingredient in skin aging therapy.
University, Taichung, Taiwan

Correspondence KEYWORDS
Wei-Wen Kuo, Department of Biological fermented flower extract of Jasminum sambac, H2O2, human skin fibroblast HS68 cells,
Science and Technology, College of
reactive oxygen species, ultraviolet B
Biopharmaceutical and Food Sciences, China
Medical University, No. 91, Hsueh-Shih Road,
Taichung 40402, Taiwan.
Email: [email protected]

Funding information
Academia-Industry Cooperation, Grant/Award
Number: 10742644; China Medical University,
Grant/Award Number: DMR-108-139

Environmental Toxicology. 2020;e23065. wileyonlinelibrary.com/journal/tox © 2020 Wiley Periodicals LLC. 1 of 13

https://doi.org/10.1002/tox.23065
2 of 13 HO ET AL.

1 | I N T RO DU CT I O N bound to Kelch-like ECH-associated protein 1 (Keap-1) and retained

in the cytosol. However, stress conditions disrupt the Nrf2/Keap-1
Changes in the skin are the most prominent features of aging. Aging interaction, resulting in Nrf2 ubiquitination and degradation.13 Upon
of the human skin is a progressive and complicated process that exposure to stress signals, Nrf2 activation facilitates its nuclear trans-
includes intrinsic aging and extrinsic aging. Extrinsic aging occurs due location to activate the expression of antioxidant genes, including
to exposure to environmental factors, such as ultraviolet heme oxygenase-1 (HO-1), γ-glutamate-cysteine synthetase (γ-GCSc)
(UV) irradiation and heavy metal pollution. Among these factors, UVB and glutathione S-transferase (GST).14 These antioxidant genes may
1
exposure is the leading factor in extrinsic aging. UV irradiation results protect dermal cells against UV-induced oxidative stress and cellular
in excessive reactive oxygen species (ROS) production and triggers a apoptosis. Therefore, natural compounds that can activate Nrf2 and
series of signal transductions, leading to dermal dysfunction, collagen its downstream antioxidant genes may be an important cellular strat-
2
loss, and wrinkle formation. egy to prevent UV-induced skin oxidative damage.
Dermal fibroblasts are the most prominent cell type in the dermis The use of probiotics and plant extracts in skin care products has
and play a crucial role in the progression of skin aging. Dermal fibro- gradually increased because of their antiaging, anti-inflammatory and
blasts can secrete large amounts of extracellular matrix (ECM) pro- antioxidant properties.15 Probiotics are known to benefit human
teins, such as collagen, hyaluronic acid (HA), and elastic fibers, that are health by modulating the intestinal immune system to protect against
responsible for conferring strength, maintaining skin moisture and harmful bacterial invasion and intestinal inflammation.16 A previous
3
providing elasticity. Collagen is the most abundant protein in animals report has shown that probiotics alone or in combination with plant
and the major structural protein in the dermal ECM. Procollagen, extracts can protect against UV-induced skin cutaneous damage,
which is the precursor of collagens such as collagen I and III, is synthe- chronic skin inflammation, and atopic dermatitis.17-19 Moreover, Lac-
4
sized by dermal fibroblasts and secreted by cells in its mature form. tobacillus rhamnosus exhibits anticancer, anti-inflammation, and anti-
The amount of collagen produced is regulated by mitogen-activated melanogenisis properties in skin cells.20-22 Jasmine (Jasminum
protein kinases (MAPKs), including Jun-N-terminal kinase (JNK), sambac), an evergreen shrub, is distributed in tropical Asia. The roots,
extracellular-regulated protein kinase (ERK) and p38 kinase, which leaves, and flowers of jasmine are widely used in folk medicine. The
activate the downstream transcription factor activator protein-1 (AP- flowers of J. sambac exhibit antimicrobial, antioxidant, anticancer, and
1).5 The nuclear translocation of AP-1 upregulates matrix antidepressant properties.23,24 Furthermore, the essential oil of
metalloproteinase (MMP) expression and results in elastin and colla- J. sambac flowers possesses antibacterial activities.25 However, no
6
gen fragmentation or degradation. Furthermore, there are several studies have investigated the role of the cultured broth of
types of MMPs, such as MMP-1, MMP-3, and MMP-9, that are L. rhamnosus -fermented flower extracts of J. sambac (F-FEJS) in regu-
involved in collagen breakdown.7 Therefore, natural compounds that lating UVB/H2O2-induced dermal cell aging.
have MAPK or MMP inhibitory properties possess the potential to
prevent aging of the skin.
In addition to promoting collagen degradation, UV irradiation has 2 | M A T E R I A L S A N D M ET H O D S
been reported to affect collagen synthesis in human dermal fibroblast
cells.8 Transforming growth factor-β (TGF-β) activates its signal trans- 2.1 | Cell culture and treatments
duction pathway by specifically binding to the cell surface serine/thre-
onine kinase receptor complex, which includes TGF-type I receptor The human foreskin fibroblast cell line HS68 was obtained from the
(TGFR-I) and TGF-type II receptor (TGFR-II).9 The binding of TGF-β to Bioresourse Collection and Research Center (BCRC, Hsinchu, Taiwan).
its receptors leads to the phosphorylation of Smad2 and Smad3, The HS68 cells were grown in Dulbecco's modified Eagle's medium
10
which in turn form heteromeric complexes with Smad4. Activated (Sigma-Aldrich D5796, Missouri) containing 2 mM L-glutamine sup-
Smad2/3/4 complexes then translocate to the nucleus and transcrip- plemented with 10% HyClone fetal bovine serum (Logan, Utah),
tionally regulate collagen type I (COL1A1) and collagen type III 100 μg/mL penicillin, and 100 μg/mL streptomycin in humidified air
(COL3A1) expression by binding to the Smad-binding elements (SBEs) (5% CO2) at 37 C. All the experiments were conducted using low-
in promoters. Previous studies have reported that UV-B irradiation passage (passage <30) human skin fibroblasts. For H2O2 (Sigma,
impairs the TGF-β/Smad pathway by decreasing TGF-β expression, St. Louis, Missouri) treatment, the cells were incubated with 200 μM
thus inhibiting Smad2/3/4 complex nuclear translocation and leading H2O2 for 1 hour and then cotreated with different concentrations of
to reduced collagen synthesis.11 Taken together, the upregulation of F-FEJS for 23 hours. Before UVB exposure, the cells were washed
the components of the TGF-β/SMAD collagen synthesis pathways with PBS and placed in 1 mL fresh PBS. For UVB irradiation, the cells
may be an effective therapeutic approach to protect against UV- were exposed to a crosslinker (CX-2000, Upland, California) and UVB
induced dermal collagen degradation. bulbs at an intensity of 40 mJ/cm2. After UVB exposure, the cells
Oxidative stress is the most prominent cause of skin aging. A key were incubated with different concentrations of F-FEJS for 24 hours.
mechanism in cellular defenses against oxidative stress is the activa- The cells were exposed to 40 mJ/cm2 UVB or 200 μM H2O2 because
tion of the NF-E2-related nuclear factor 2 (Nrf2)/antioxidant response we previously demonstrated that these doses could induce HS68 der-
element (ARE) signaling pathway.12 Under normal conditions, Nrf2 is mal cell senescence but not apoptosis.26,27
HO ET AL. 3 of 13

2.2 | Antibodies and reagents 2.5 | Intracellular reactive oxygen species

production
The following antibodies were used in this study: anti-COL1A1 (sc-
28 657, Santa Cruz, California), anti-COL3A1 (sc-271 249, Santa Cruz, The production of intracellular ROS in response to UVB or H2O2
California), anti-ERK1 (K-23) (sc-94, Santa Cruz, California), anti-p-ERK exposure and F-FEJS treatment was examined by flow cytometry and
1/2 (E-4) (sc-7383, Santa Cruz, California), anti-GAPDH (sc-32 233, fluorescence microscopy using DCFH-DA (D6883, Sigma). After
Santa Cruz, California), anti-HDAC1 (c-19) (sc-6298, Santa Cruz, Califor- removing the culture medium, the cells were washed twice with PBS
nia), anti-HO-1 (H-105) (sc-10 789, Santa Cruz, California), anti-JNK (sc- and incubated with 5 μM DCFH-DA in PBS at 37 C for 30 minutes.
571, Santa Cruz, California), anti-p-JNK (sc-6254, Santa Cruz, California), The DCF fluorescence was measured by a BD FACS Canto M II flow
anti-c-Jun (G-4) (sc-74 543, Santa Cruz, California), anti-p-c-Jun cytometer (Becton-Dickinson, Franklin Lakes, New Jersey) and fluo-
(Ser63/73) (sc-16 312-R, Santa Cruz, California), anti-MMP-1 (3B6) (sc- rescence microscope (Olympus DP73, Tokyo, Japan) in a detection
21 731, Santa Cruz, California), anti-Nrf2 (ab89443, Abcam), anti-p-Nrf2 range of 515 to 565 nm (green). The analysis was repeated three
(S40) (ab76026, Abcam, Cambridge, UK), anti-P16INK4A (10883-1-AP, times for each treatment to perform statistical analysis.
Proteintech, Manchester, UK), anti- p21 (F-5) (sc-6246, Santa Cruz, Cali-
fornia), anti-p53 (1C12) (#2524, Cell Signaling, Danvers, Massachusetts),
anti-p38α (A1F7) (sc-33 688, Santa Cruz, California), anti-p-p38 (D-8) 2.6 | Luciferase and fluorescence reporter assays
(sc-7973, Santa Cruz, California), anti-p-Smad2/3 (Ser433/435) (sc-
11 769, Santa Cruz, California), anti-Smad2 (ab63576, Abcam), anti- The pGL3-ARE plasmid was a gift from Dr Being-Sun Wung
Smad3 (sc-8332, Santa Cruz, California), TGFβ1 (C-16) (sc-31 609, Santa (Department of Microbiology, Immunology and Biopharmaceuticals,
Cruz, California) and anti-β-actin (sc-47 778, Santa Cruz, California). All National Chiayi University). The ARE transcriptional activity was mea-
the secondary antibodies, anti-rabbit (A0545), anti-mouse (A9044), and sured using the Dual-Luciferase Reporter (DLR) Assay System
anti-goat (A5420), were purchased from Santa Cruz, California. (Promega, California). HS68 cells (15 × 104 cells/well) were seeded in
6-well plates and grown until they reached 80% confluence. The
HS68 cells were then transfected with the pGL3-ARE luciferase
2.3 | Preparation of fermentation broth of jasmine reporter constructs or PGL3 empty vector using the PureFection
flower extracts transfection reagent according to the manufacturer's instructions
(System Biosciences, California). After 1 hour of plasmid transfection,
For the current study, L. rhamnosus was purchased from ATCC the cells were treated with different concentrations of F-FEJS for
(53103). Jasmine flower extracts were purchased from the Zen Heart 23 hours. Briefly, 100 μL of luciferase substrate was added to 20 μL
Group Biopharmaceutical. The jasmine flower extracts (0.2 g) were of lysate, and the luciferase activity was measured using an LB940
fermented with L. rhamnosus (2.2 x 107 cfu/mL) by incubation at 37 C Multilabel Reader (Berthold Technologies, Bad Wildbad, Germany).
for 48 hours in a medium consisting of 15 g/L tryptone, 2.5 g/L meat The analysis was repeated three times for each treatment to perform
extract, 7.5 g/L yeast extract, 4.5 g/L K2HPO4, 0.05 g/L cysteine HCl, statistical analysis.
2.5 g/L lactose, 7.5 g/L glucose and 80 mL/L Tween. The medium was
adjusted to pH 6.5 ± 0.1 for fermentation. The fermented broth was
then centrifuged at 8000 ×g for 25 minutes, and the supernatant was 2.7 | Western blot analysis
filtered using a 0.45-μm membrane filter. The filtrate of the cultured
broth was stored in the dark at 4 C. Western blot analysis was performed following the protocols
described in previous reports with slight modifications.28 The total
protein fractions were isolated by mixing with lysis buffer (50 mM
2.4 | MTT [3-(4,5-dimethylthiazol-2-yl)- Tris, pH 7.5, 0.5 M NaCl, 1.0 mM EDTA, pH 7.5, 10% glycerol, 1 mM
2,5-diphenyltetrazolium-bromide] assay BME, 1% IGEPAL-630, and protease inhibitor cocktail) and incubating
on ice for 1 hour. After incubation, the supernatant was collected by
HS68 cells were seeded in 24-well plates (2 × 105 cells/well), exposed to centrifugation at 12 000 g for 15 minutes at 4 C. The protein concen-
UVB or H2O2 and then cotreated with different concentrations of F- tration was determined using the Bradford method. After protein
FEJS for 24 hours. The cell viability was determined using the MTT quantification, the samples were heated at 95 C for 10 minutes. Equal
assay. After washing twice with PBS, the cells were cultured with 100 μL amounts of the denatured proteins (20 μg) were loaded and separated
of 0.5 mg/mL MTT solution in phenol red-free DMEM. After 4 hours of using 6% to 12% SDS-PAGE and then transferred onto PVDF mem-
incubation at 37 C, the medium was removed and then incubated with branes (Millipore, Belford, Massachusetts). The membranes were
100 μL dimethyl sulfoxide (DMSO) with shaking for 10 minutes. The blocked with blocking buffer (5% nonfat dry milk, 20 mM Tris-HCl,
absorbance of the formazan product was measured at 570 nm using a pH 7.6, 150 mM NaCl, and 0.1% Tween 20) for at least 1 hour at
spectrophotometer reader. The analysis was repeated three times for RT. The membranes were then incubated with primary antibodies in
each treatment to perform statistical analysis. the solution described above on an orbital shaker at 4 C overnight.
4 of 13 HO ET AL.

Following the primary antibody incubation, the membranes were incu- results were quantified using ImageJ software and were processed using
bated with horseradish peroxidase-conjugated secondary antibodies the Adobe Photoshop.
(anti-rabbit, anti-mouse, or anti-goat IgG) diluted 80 000-fold in the
solution described above for 1 hour. The blots were imaged using an
ImageQuant LAS4000 mini (GE Healthcare Life Sciences, Little 3 | RE SU LT S
Chalfont, UK) with a chemiluminescence substrate.
3.1 | F-FEJS inhibits UVB/H2O2-induced
cytotoxicity in HS68 human dermal fibroblast cells
2.8 | Isolation of nuclear and cytoplasmic extracts
To ensure the safety of the drugs, human dermal fibroblast (HS68)
Nuclear extracts were used to analyze p-Smad2/3, p-c-jun and p-Nrf2 cells were treated with different percentages of F-FEJS, and an MTT
expression. A nuclear/cytosol fractionation kit (Catalog #K266-25, assay was used to assess cell viability. Treatment with different con-
BioVision) was used to separate the nuclear and cytosolic proteins centrations (0%-2.5%, 24 hours) of F-FEJS led to no cytotoxicity in
according to the manufacturer's instructions. After carefully aspirating the HS68 cells (Figure 1A). After exposure to H2O2 (200 μM) or UVB
the supernatant, the cells were resuspended in 200 μL of ice-cold (40 mJ/cm2), cell viability was remarkably decreased to approximately
CEB-A mix (proteinase inhibitor cocktail, DTT), vigorously vortexed at 60%. At 0% to 0.25%, F-FEJS increased cell viability in a dose-
the highest speed for 14 seconds, and incubated on ice for 10 minutes. dependent manner, but decreased cell viability at 5%, in UVB/H2O2-
Eleven microliters of ice-cold CEB-B was added to the tube. After vig- treated HS68 cells (Figure 1B,C). Therefore, 2.5% F-FEJS was reg-
orously vortexing for 5 seconds, the samples were incubated on ice arded as the safe dose in this study.
for 1 minute and then centrifuged at 16 000 g for 5 minutes at 4 C.
Then, the supernatant (cytoplasmic fraction) was carefully aspirated,
and the pellet was resuspended in 100 μL of ice-cold NEB mix (pro- 3.2 | F-FEJS attenuates UVB/H2O2-induced
teinase inhibitor cocktail, DTT) and vigorously vortexed for intracellular ROS production in HS68 cells
15 seconds every 10 minutes for a total of 40 minutes. After
vortexing, the suspension was centrifuged at 16000 g for 10 minutes The increased cell cytotoxicity in response to UVB/H2O2 exposure
at 4 C. The supernatants (nuclear extracts) were stored at −80 C for has been reported to be associated with ROS formation in human skin
further use. dermal fibroblast cells.29 Therefore, we examined the ROS-scavenging
abilities of F-FEJS in HS68 cells. The fibroblast cells were exposed to
H2O2 (200 μM) or UVB (40 mJ/cm2) and treated with 2.5% F-FEJS for
2.9 | Senescence-associated β-galactosidase (SA- 24 hours. DCF-DA staining was used to measure intracellular ROS
β-gal) staining generation and analyzed by fluorescence microscopy or flow cyto-
metry. The fluorescence staining and flow cytometry results indicated
The SA-β-gal staining kit was used to detect senescent HS68 cells that treatment with 2.5% F-FEJS protected HS68 cells from H2O2- or
(#9860, Cell Signaling, Danvers, Massachusetts). HS68 fibroblast cells UVB-induced ROS generation (Figure 2A-C).
were seeded in 4-well slides followed by UVB/H2O2 exposure and F-
FEJS treatment. The culture medium was removed, and the cells were
washed twice with ice-cold PBS. Fixation solution was added and 3.3 | Antiaging effects of F-FEJS on UVB/H2O2-
incubated at RT for 10 minutes. The cells were then washed three induced senescence of HS68 cells
times with ice-cold PBS and incubated with staining solution over-
night. After washing, the cells were photographed under a light micro- Based on the results of ROS inhibition by F-FEJS, we next tested
scope (Olympus DP73, Tokyo, Japan). whether F-FEJS would protect fibroblast cells from UVB- or H2O2-
induced aging. HS68 fibroblast cells were incubated with H2O2 (200 μM)
or UVB (40 mJ/cm2) for 24 hours to induce premature senes-
2.10 | Statistical analysis cence.26,27,30 The levels of aging markers, such as p16, p21, and p53,
were significantly upregulated in the H2O2− or UVB-exposed cells. How-
The statistical analyses were performed using the GraphPad Prism 5 sta- ever, treatment with F-FEJS (0.1%, 0.5%, and 2.5%) led to a dose-
tistical software. Overall significant differences between the means of dependent reduction in these aging-related markers (Figure 3A). In addi-
multiple groups are assessed by analysis of variance. The data were ana- tion, the number of SA-β-Gal-positive cells was higher in the cells treated
lyzed by separate one-way ANOVA. Significant differences between the with UVB or H2O2, while F-FEJS (0.1% and 2.5%) treatment reduced the
individual means were determined by Tukey's tests. The quantitative number of SA-β-Gal-positive cells (Figure 3B). Taken together, the sup-
data are shown as the mean ± SE corresponding to three or more repli- pression of senescence by F-FEJS suggested that it may exert dermo-
cates. A P-value of <.05 was considered statistically significant. All the protective effects against UVB- or H2O2-induced damage.
HO ET AL. 5 of 13

F I G U R E 1 Fermented flower extract of Jasminum sambac (F-FEJS) inhibits UVB/H2O2-induced cytotoxicity in HS68 human dermal fibroblast
cells. (A) HS68 cells were treated with different concentrations of F-FEJS (0, 0.05, 0.10, 0.25, 0.50, 1.0, 2.5, and 5.0%) for 24 hours. (B) HS68 cells
were exposed to H2O2 (200 μM) for 1 hour and then cotreated with different percentages of F-FEJS as indicated for 23 hours. (C) Before UVB
(40 mJ/cm2) exposure, the cells were washed with PBS and placed in 1 mL fresh PBS. The cells were then incubated with F-FEJS at the indicated
concentrations for 24 hours. The cytoprotective effects of F-FEJS were determined by the MTT assay. The control cells were set to 100%
viability. The values shown are the means ± SE. Quantification of the results is shown (n = 3) **P < .01, ***P < .001 vs untreated control cells;
#
P < .05, ##P < .01, ###P < .001 vs H2O2/UVB-treated cells

F I G U R E 2 Fermented flower extract of Jasminum sambac (F-FEJS) attenuates UVB/H2O2-induced intracellular reactive oxygen species (ROS)
production in HS68 cells. HS68 cells were exposed to H2O2 (200 μM) for 1 hour and then cotreated with 2.5% F-FEJS for 23 hours. Before UVB
(40 mJ/cm2) exposure, the cells were washed with PBS, and 1 mL fresh PBS was added. The cells were then incubated with F-FEJS at the
indicated concentrations for 24 hours. (A, B) The intracellular ROS levels were determined with 2',7'–dichlorofluorescin diacetate (DCF-DA)
(5 μM) by using fluorescence microscopy and flow cytometry. (C) Quantification results of % of DCF fluorescence from Figure 2B. The values
shown are the means ± SE. Quantification of the results is shown (n = 3) ***P < .001 vs untreated control cells; ###P < .001 vs H2O2/UVB-treated
cells [Color figure can be viewed at wileyonlinelibrary.com]
6 of 13 HO ET AL.

FIGURE 3 Legend on next page.

HO ET AL. 7 of 13

F I G U R E 4 Inhibitory effects of fermented flower extract of Jasminum sambac (F-FEJS) on UVB/H2O2-induced collagen degradation in HS68
cells. HS68 cells were exposed to H2O2 (200 μM) for 1 hour and then cotreated with (0.1, 0.5, and 2.5%) F-FEJS for 23 hours. Before UVB
(40 mJ/cm2) exposure, the cells were washed with PBS and placed in 1 mL fresh PBS. The cells were then incubated with F-FEJS at the indicated
concentrations for 24 hours. (A, B) Collagen synthesis-related pathway components, such as TGF-β, p-smad2/3, smad2, smad3, COL1A1,
COL3A1, and collagen degradation-related protein MMP-1, were detected by western blot. GAPDH was used as the loading control. The values
shown are the means ± SE. Quantification of the results is shown (n = 3) *P < .05, **P < .01, ***P < .001 vs untreated control cells; #P < .05,
##
P < .01, ###P < .001 vs H2O2/UVB-treated cells

F I G U R E 3 Antiaging effects of Fermented flower extract of Jasminum sambac (F-FEJS) on UVB/H2O2-induced senescence in HS68 cells.
HS68 cells were exposed to H2O2 (200 μM) for 1 hour and then cotreated with (0.1, 0.5, and 2.5%) F-FEJS for 23 hours. Before UVB (40 mJ/cm2)
exposure, the cells were washed with PBS and placed in 1 mL fresh PBS. The cells were then incubated with F-FEJS at the indicated
concentrations for 24 hours. (A, B) The protein levels of p53, p21, and p16 were detected by western blot. GAPDH was used as the loading
control. The values shown are the means ± SE. Quantification of the results is shown (n = 3) *P < .05, **P < .01, ***P < .001 vs untreated control
cells; #P < .05, ##P < .01, ###P < .001 vs H2O2/UVB-treated cells. (C, D) After treatment, the cells were stained with an SA-β-Gal Staining Kit (Cell
Signaling, #9860) to detect the senescence-associated β-gal activity. The positive cells were stained green, indicating senescent cells [Color figure
can be viewed at wileyonlinelibrary.com]
8 of 13 HO ET AL.

3.4 | Inhibitory effects of F-FEJS on UVB/H2O2- 3.5 | Effects of F-FEJS on UVB/H2O2-induced

induced collagen degradation in HS68 cells MAPK and c-jun activation in HS68 cells

To elucidate the effects of F-FEJS on UVB/H2O2-induced collagen In a previous study, MAPK/c-jun activation promoted collagen break-
degradation, we assessed TGFβ/SMAD (collagen synthesis pathways) down by increasing MMP expression in dermal fibroblasts.11 To deter-
and MMP-1 (collagenase) expression in UVB- or H2O2-exposed cells. mine whether F-FEJS could inhibit MMP-1 expression by blocking
Treatment with F-FEJS (0.1%, 0.5%, and 2.5%) led to the inhibition of MAPK and c-jun activation, we examined the levels of phosphorylated
MMP-1 and enhancement of the TGFβ, p-smad2/3, COL1A1, and ERK, JNK, P38 and c-jun. Treatment with F-FEJS (0.1%, 0.5%, and
COL3A1 protein levels in UVB- or H2O2-exposed cells in a dose- 2.5%) led to decreases in the p-ERK, p-JNK, p-P38, and p-c-jun pro-
dependent manner (Figure 4A,B). Our results provide evidence that F- tein levels in UVB- or H2O2-exposed cells in a dose-dependent man-
FEJS can protect UVB- or H2O2-exposed dermal fibroblasts from col- ner (Figure 5A,B). These findings indicate that the upregulation of
lagen degradation by upregulating the components of the TGFβ/ collagen synthesis by F-FEJS was accompanied by the inhibition of
SMAD collagen synthesis pathways. MAPK and c-jun activation.

F I G U R E 5 Effects of fermented flower extract of Jasminum sambac (F-FEJS) on UVB/H2O2-induced MAPK and c-jun activation in HS68 cells.
HS68 cells were exposed to H2O2 (200 μM) for 1 hour and then cotreated with (0.1, 0.5, and 2.5%) F-FEJS for 23 hours. Before UVB (40 mJ/cm2)
exposure, the cells were washed with PBS and placed in 1 mL fresh PBS. The cells were then incubated with F-FEJS at the indicated
concentrations for 24 hours. (A, B) The protein levels of ERK, p-ERK, p38, p-p38, JNK, p-JNK, c-jun, and p-c-jun were detected by western blot.
GAPDH was used as the loading control. The values shown are the means ± SE. Quantification of the results is shown (n = 3) *P < .05, **P < .01,
***
P < .001 vs untreated control cells; #P < .05, ##P < .01, ###P < .001 vs H2O2/UVB-treated cells
HO ET AL. 9 of 13

3.6 | F-FEJS enhances Nrf2/ARE activation and shown in Figure 6A,B, treatment with F-FEJS (0.1%, 0.5%, and 2.5%)
upregulates antioxidant gene expression (HO-1) in enhanced the p-Nrf2 and HO-1 levels in UVB- or H2O2-exposed
HS68 cells cells (Figure 6A,B). It has been reported that after nuclear transloca-
tion, Nrf2 binds to AREs in the promoter regions of several antioxi-
Nrf2 is a well-known transcriptional regulator that modulates antiox- dant genes.32 Next, we transfected a luciferase vector carrying ARE
idant genes, such as HO-1, and plays a critical role in the cyto- sequences. The data showed that F-FEJS (0.1%, 0.25%, 0.5%, 1.0%,
31,32
protective effects against ROS. Therefore, we assume that the and 2.5%) treatment increased the ARE luciferase activities in a
cytoprotective effects of F-FEJS against UVB or H2O2 exposure- dose-dependent manner (Figure 6C). These findings demonstrate
mediated damage may be due to Nrf2/ARE signaling activation. As that the induction of HO-1 occurs through Nrf2/ARE signaling.

F I G U R E 6 Fermented flower extract of Jasminum sambac (F-FEJS) enhances Nrf2/ARE activation and upregulates antioxidant gene
expression (HO-1) in HS68 cells. HS68 cells were exposed to H2O2 (200 μM) for 1 hour and then cotreated with (0.1, 0.5, and 2.5%) F-FEJS for
23 hours. Before UVB (40 mJ/cm2) exposure, the cells were washed with PBS and placed in 1 mL fresh PBS. The cells were then incubated with
F-FEJS at the indicated concentrations for 24 hours. (A, B) The protein levels of Nrf2, p-Nrf2, and HO-1 were detected by western blot. GAPDH
was used as the loading control. (C) HS68 cells were transfected with ARE-luciferase or PGL3 promoter plasmid (PGL) for 1 hour. The cells were
then treated with different percentages of F-FEJS (0, 0.10, 0.25, 0.50, 1.0, and 2.5%) for 23 hours, and the luciferase activity was analyzed. The
values shown are the means ± SE. Quantification of the results is shown (n = 3) *P < .05, ***P < .001 vs only ARE plasmid-treated cells
10 of 13 HO ET AL.

F I G U R E 7 Effects of fermented flower extract of Jasminum sambac (F-FEJS) on the nuclear translocation of p-c-jun, p-smad2/3, and p-Nrf2
in HS68 cells exposed to H2O2/UVB. (A, B) HS68 cells were exposed to H2O2 (200 μM) for 1 hour and then cotreated with (0.1, 0.5, and 2.5%) F-
FEJS for 23 hours. Before UVB (40 mJ/cm2) exposure, the cells were washed with PBS and placed in 1 mL fresh PBS. The cells were then
incubated with F-FEJS at the indicated concentrations for 24 hours. The protein levels of p-c-jun, p-smad2/3, and p-Nrf2 in the cytosolic and
nuclear fractions were estimated by western blot. GAPDH and HDAC1 acted as the internal controls for the cytosolic and nuclear fractions,
respectively [Color figure can be viewed at wileyonlinelibrary.com]

F I G U R E 8 Schematic summarizing the inhibition of UVB/H2O2-induced skin aging by fermented flower extract of Jasminum sambac (F-FEJS)
marked with red arrows in HS68 human dermal fibroblast cells [Color figure can be viewed at wileyonlinelibrary.com]
HO ET AL. 11 of 13

3.7 | Effects of F-FEJS on the nuclear translocation but this increase was reversed following 2.5% F-FEJS treatment,
of p-c-jun, p-smad2/3, and p-Nrf2 in HS68 cells suggesting that F-FEJS may possess ROS scavenging abilities
exposed to UVB or H2O2 (Figure 2).
In response to UVB or H2O2, which are cytotoxic to dermal fibro-
To demonstrate the activation of transcription factors by F-FEJS dur- blasts, cells can exert certain self-defense actions. Nrf2 plays an
ing UVB or H2O2 exposure, we performed nuclear and cytosolic sepa- important role in maintaining cellular redox balance, particularly in
ration to examine the activation of c-jun, smad2/3 and Nrf2. The response to oxidative stress, by activating a variety of defensive pro-
translocation of p-Nrf2 was slightly higher in UVB- or H2O2-exposed teins.40 Accumulating evidence suggests that Nrf2 signaling pathways
cells, while treatment with different doses of F-FEJS (0.1%, 0.5%, and provide cellular protection against UV- or H2O2-induced oxidative
2.5%) further enhanced p-Nrf2 nuclear translocation. Furthermore, damage in skin keratinocytes and fibroblasts.32,41,42 Our results dem-
treatment with F-FEJS resulted in the downregulation of p-c-jun and onstrated that Nrf2 accumulates in the nucleus and that downstream
the upregulation of p-smad2/3 in the nuclear fractions of UVB- or antioxidant genes are slightly increased in HS68 cells exposed to
H2O2-exposed dermal fibroblast cells (Figure 7A,B). Taken together, UVB/H2O2. This finding is similar to our previous study. In either high
our results confirmed that the Nrf2, smad2/3 and c-jun pathways glucose-induced cardiac cell death or H2O2-induced dermal fibroblast
were the three major pathways involved in the attenuation of UVB- cell damage models, this stress eventually leads to ROS formation and
or H2O2-induced senescence, ROS production and collagen degrada- Nrf2 activation.26,43,44 Interestingly, treatment with F-FEJS further
tion in F-FEJS-treated HS68 human dermal fibroblast cells (Figure 8). enhanced antioxidant responses to protect dermal cells from UVB- or
H2O2-induced damage (Figure 6).
Collagen degradation can weaken the stability of the ECM,
4 | DISCUSSION resulting in skin aging.45 MMP-1, which acts as collagenase, is
secreted in response to stimulation by ROS, UV radiation or cyto-
We utilized UVB irradiation and H2O2 treatment to induce ROS gen- kines.46 The promoter regions of MMPs contain AP-1-binding sites,
eration in dermal fibroblast cells. When our skin is exposed to air in and AP-1 is directly downstream of MAPKs, indicating that MAPKs
the environment, it is irradiated by UV light, which consists of approx- could regulate collagen degradation. Furthermore, COL1A1 and
imately 95% UVA (315-400 nm) and only approximately 5% UVB COL2A1 were downregulated by enhancing AP-1 expression.6 Several
(280-315 nm).33 UV irradiation causes the formation of ROS, includ- reports demonstrate that AP-1 can downregulate procollagen genes
• −
ing superoxide ( O2 ), hydrogen peroxide (H2O2) and hydroxyl radical by suppressing TGF-β/SMAD signaling pathways.47 Therefore,
• 34
( OH). Oxidative stress, in turn, can induce DNA damage, resulting inhibiting MAPK/AP-1/MMP activation or promoting collagen synthe-
in skin senescence and inflammation and accelerating malignant tumor sis may be potential mechanisms to prevent skin aging. In the current
formation.35,36 Although UVA is the most abundant in sunlight, UVB study, our findings demonstrated that F-FEJS inhibited UVB/H2O2-
is the main factor that leads to UV-induced skin photodamage due to induced MAPK/AP-1/MMP activation and restored collagen expres-
its high energy. Previous studies reported that UVA irradiation can sion by activating the TGF-β/SMAD pathways (Figures 4, 5, 7).
induce minimal damage to the skin epidermis, and it also takes longer In this study, our results showed that F-FEJS was able to amelio-
to induce severe damage, while UVB irradiation can result in epider- rate UVB- or H2O2-induced dermal fibroblast cell aging by promoting
mal and dermal changes, such as hyperpigmentation, hyperplasia and cell survival, TGF-β/SMAD collagen synthesis pathways, and Nrf2/
psoriasis.37,38 ARE antioxidant signaling pathways and by suppressing MAPK/AP-1/
Various studies have indicated that low doses of UVB or H2O2 MMP activation (Figure 8). Our results indicated that the broth of the
result in dermal fibroblast cell aging rather than apoptosis.39 In this flower extract of J. sambac fermented with L. rhamnosus may be a
study, doses of H2O2 (200 μM) and UVB (40 mJ/cm ) that induce cell
2
potential therapeutic candidate for use in cosmetic agents to protect
aging were chosen. First, we found that F-FEJS treatment did not against UV-induced skin photoaging. Further study will investigate
cause any cytotoxicity (0%-2.5%) but rather substantially increased the protective effects of F-FEJS against UVB irradiation-induced dam-
cell viability in UVB- and H2O2-treated dermal fibroblast cells age in animal models.
(Figure 1). Notably, aging marker expression and SA-β-gal activity
were suppressed by F-FEJS treatment in a dose-dependent manner ACKNOWLEDG MENTS
(Figure 3), indicating that F-FEJS could play a protective role as an This work was supported by China Medical University (DMR-
antiaging agent in human dermal fibroblasts exposed to UVB 108-139) and Academia-Industry Cooperation (10742644). We thank
and H2O2. Dr Being-Sun Wung (Department of Microbiology, Immunology and
The molecular mechanism of skin senescence is very complex. Biopharmaceuticals, National Chiayi University, Chiayi, Taiwan) pro-
Several studies indicate that UV-induced ROS formation may regulate vided the pGL3-ARE plasmid in this study.
various important signaling pathways, which subsequently lead to the
activation and nuclear translocation of transcription factors, including OR CID
AP-1, NF-ĸB, and Nrf2.7 Our results showed that UVB and H2O2 Chih-Yang Huang https://orcid.org/0000-0003-2347-0411
increased the production of intracellular ROS in dermal fibroblast cells, Wei-Wen Kuo https://orcid.org/0000-0002-2097-2240
12 of 13 HO ET AL.

RE FE R ENC E S photoprotective agent against UV-induced carcinogenesis. Br J Nutr.

1. Chung JH. Photoaging in Asians. Photodermatol Photoimmunol Photo- 2013;109(3):457-466.
med. 2003;19(3):109-121. 23. Talib WH, Mahasneh AM. Antiproliferative activity of plant extracts
2. Uitto J. Understanding premature skin aging. N Engl J Med. 1997;337 used against cancer in traditional medicine. Sci Pharm. 2010;78(1):
(20):1463-1465. 33-45.
3. Quan T, Fisher GJ. Role of age-associated alterations of the dermal 24. Abdoul-Latif FM, Mohamed N, Edou P, et al. Antimicrobial and anti-
extracellular matrix microenvironment in human skin aging: a mini- oxidant activities of essential oil and methanol extract of Matricaria
review. Gerontology. 2015;61(5):427-434. chamomilla L. from Djibouti. J Med Plant Res. 2011;5(9):1512-1517.
4. Fisher GJ, Wang ZQ, Datta SC, Varani J, Kang S, Voorhees JJ. Patho- 25. Rath CC, Devi S, Dash SK, Mishra RK. Antibacterial potential assess-
physiology of premature skin aging induced by ultraviolet light. N Engl ment of jasmine essential oil against E. coli Indian. J Pharm Sci. 2008;
J Med. 1997;337(20):1419-1428. 70(2):238-241.
5. Bachelor MA, Bowden GT. UVA-mediated activation of signaling 26. Wen SY, Chen JY, Weng YS, et al. Galangin suppresses H2 O2
pathways involved in skin tumor promotion and progression. Semin -induced aging in human dermal fibroblasts. Environ Toxicol. 2017;32
Cancer Biol. 2004;14(2):131-138. (12):2419-2427.
6. Bornstein P. Regulation of expression of the alpha 1 (I) collagen gene: 27. Wen SY, Chen JY, Chen CJ, Huang CY, Kuo WW. Protective effects
a critical appraisal of the role of the first intron. Matrix Biol. 1996;15 of galangin against H2 O2 -induced aging via the IGF-1 signaling
(1):3-10. pathway in human dermal fibroblasts. Environ Toxicol. 2020;35(2):
7. Chiang HM, Chen HC, Chiu HH, Chen CW, Wang SM, Wen KC. 115-123.
Neonauclea reticulata (Havil.) Merr stimulates skin regeneration after 28. Wang ZH. Anti-glycative effects of asiatic acid in human keratinocyte
UVB exposure via ROS scavenging and modulation of the cells. Biomedicine. 2014;4:19.
MAPK/MMPs/collagen pathway. Evid Based Compl Alternat Med. 29. Patwardhan J, Bhatt P. Ultraviolet-B protective effect of flavonoids
2013;2013:324864. from Eugenia caryophylata on human dermal fibroblast cells. Pharma-
8. Quan T, He T, Kang S, Voorhees JJ, Fisher GJ. Solar ultraviolet irradia- cogn Mag. 2015;11(3):S397-S406.
tion reduces collagen in photoaged human skin by blocking trans- 30. Chen JH, Ozanne SE, Hales CN. Methods of cellular senescence
forming growth factor-beta type II receptor/smad signaling. induction using oxidative stress. Methods Mol Biol. 2007;371:
Am J Pathol. 2004;165(3):741-751. 179-189.
9. Quan T, He T, Kang S, Voorhees JJ, Fisher GJ. Ultraviolet irradiation 31. Kobayashi A, Kang MI, Watai Y, et al. Oxidative and electrophilic
alters transforming growth factor beta/smad pathway in human skin stresses activate Nrf2 through inhibition of ubiquitination activity of
in vivo. J Invest Dermatol. 2002;119(2):499-506. Keap1. Mol Cell Biol. 2006;26(1):221-229.
10. Palladino MA, Morris RE, Starnes HF, Levinson AD. The transforming 32. Hirota A, Kawachi Y, Itoh K, et al. Ultraviolet a irradiation induces NF-
growth factor-betas. A new family of immunoregulatory molecules. E2-related factor 2 activation in dermal fibroblasts: protective role in
Ann N Y Acad Sci. 1990;593:181-187. UVA-induced apoptosis. J Invest Dermatol. 2005;124(4):825-832.
11. Rittie L, Fisher GJ. UV-light-induced signal cascades and skin aging. 33. Debacq-Chainiaux F, Leduc C, Verbeke A, Toussaint O. UV, stress
Ageing Res Rev. 2002;1(4):705-720. and aging. Dermatoendocrinol. 2012;4(3):236-240.
12. Zhang DD. Mechanistic studies of the Nrf2-Keap1 signaling pathway. 34. De Jager TL, Cockrell AE, Du Plessis SS. Ultraviolet light induced gen-
Drug Metab Rev. 2006;38(4):769-789. eration of reactive oxygen species. Adv Exp Med Biol. 2017;996:
13. Kaspar JW, Niture SK, Jaiswal AK. Nrf2:INrf2 (Keap1) signaling in oxi- 15-23.
dative stress. Free Radic Biol Med. 2009;47(9):1304-1309. 35. Bech-Thomsen N, Wulf HC. Carcinogenic and melanogenic effects of
14. Kim J, Cha YN, Surh YJ. A protective role of nuclear factor-erythroid a filtered metal halide UVA source and a tubular fluorescent UVA tan-
2-related factor-2 (Nrf2) in inflammatory disorders. Mutat Res. 2010; ning source with or without additional solar-simulated UV radiation in
690(1–2):12-23. hairless mice. Photochem Photobiol. 1995;62(4):773-779.
15. Rautava S, Isolauri E. The development of gut immune responses and 36. Oikarinen A, Karvonen J, Uitto J, Hannuksela M. Connective tissue
gut microbiota: effects of probiotics in prevention and treatment of alterations in skin exposed to natural and therapeutic UV-radiation.
allergic disease. Curr Issues Intest Microbiol. 2002;3(1):15-22. Photodermatol. 1985;2(1):15-26.
16. Parvez S, Malik KA, Ah Kang S, Kim HY. Probiotics and their fer- 37. Amaro-Ortiz A, Yan B, D'Orazio JA. Ultraviolet radiation, aging and
mented food products are beneficial for health. J Appl Microbiol. the skin: prevention of damage by topical cAMP manipulation. Mole-
2006;100(6):1171-1185. cules. 2014;19(5):6202-6219.
17. Bouilly-Gauthier D, Jeannes C, Maubert Y, et al. Clinical evidence of 38. Imokawa G, Ishida K. Biological mechanisms underlying the ultraviolet
benefits of a dietary supplement containing probiotic and carotenoids radiation-induced formation of skin wrinkling and sagging I: reduced
on ultraviolet-induced skin damage. Br J Dermatol. 2010;163(3): skin elasticity, highly associated with enhanced dermal elastase activ-
536-543. ity, triggers wrinkling and sagging. Int J Mol Sci. 2015;16(4):7753-
18. Kim HM, Lee DE, Park SD, et al. Oral administration of Lactobacillus 7775.
plantarum HY7714 protects hairless mouse against ultraviolet B- 39. Chung KW, Choi YJ, Park MH, et al. Molecular insights into SIRT1
induced photoaging. J Microbiol Biotechnol. 2014;24(11):1583-1591. protection against UVB-induced skin fibroblast senescence by sup-
19. Elazab N, Mendy A, Gasana J, Vieira ER, Quizon A, Forno E. Probiotic pression of oxidative stress and p53 acetylation. J Gerontol A Biol Sci
administration in early life, atopy, and asthma: a meta-analysis of clini- Med Sci. 2015;70(8):959-968.
cal trials. Pediatrics. 2013;132(3):666-676. 40. Yang HL, Lin MW, Korivi M, et al. Coenzyme Q0 regulates
20. Wickens K, Barthow C, Mitchell EA, et al. Effects of Lactobacillus NFkappaB/AP-1 activation and enhances Nrf2 stabilization in attenu-
rhamnosus HN001 in early life on the cumulative prevalence of aller- ation of LPS-induced inflammation and redox imbalance: evidence
gic disease to 11 years. Pediatr Allergy Immunol. 2018;29(8):808-814. from in vitro and in vivo studies. Biochim Biophys Acta. 2016;1859(2):
21. Holowacz S, Blondeau C, Guinobert I, Guilbot A, Hidalgo S, Bisson JF. 246-261.
Lactobacillus salivarius LA307 and Lactobacillus rhamnosus LA305 atten- 41. Oyewole AO, Wilmot MC, Fowler M, Birch-Machin MA. Comparing
uate skin inflammation in mice. Benef Microbes. 2018;9(2):299-309. the effects of mitochondrial targeted and localized antioxidants with
22. Weill FS, Cela EM, Paz ML, Ferrari A, Leoni J, Gonzalez Maglio DH. cellular antioxidants in human skin cells exposed to UVA and hydro-
Lipoteichoic acid from Lactobacillus rhamnosus GG as an oral gen peroxide. FASEB J. 2014;28(1):485-494.
HO ET AL. 13 of 13

42. Hseu YC, Lo HW, Korivi M, Tsai YC, Tang MJ, Yang HL. Dermato- 46. Quan T, Qin Z, Xia W, Shao Y, Voorhees JJ, Fisher GJ. Matrix-
protective properties of ergothioneine through induction of degrading metalloproteinases in photoaging. J Investig Dermatol Symp
Nrf2/ARE-mediated antioxidant genes in UVA-irradiated human Proc. 2009;14(1):20-24.
keratinocytes. Free Radic Biol Med. 2015;86:102-117. 47. Hwang KA, Yi BR, Choi KC. Molecular mechanisms and in vivo mouse
43. Tsai CY, Wang CC, Lai TY, et al. Antioxidant effects of diallyl trisulfide models of skin aging associated with dermal matrix alterations. Lab
on high glucose-induced apoptosis are mediated by the PI3K/Akt- Anim Res. 2011;27(1):1-8.
dependent activation of Nrf2 in cardiomyocytes. Int J Cardiol. 2013;
168(2):1286-1297.
44. Tsai CY, Wen SY, Cheng SY, et al. Nrf2 activation as a protective
feedback to limit cell death in high glucose-exposed cardiomyocytes. How to cite this article: Ho C-C, Ng S-C, Chuang H-L, et al.
J Cell Biochem. 2017;118(7):1659-1669. Extracts of Jasminum sambac flowers fermented by
45. Chiang HM, Chiu HH, Liao ST, Chen YT, Chang HC, Wen KC. Iso-
Lactobacillus rhamnosus inhibit H2O2- and UVB-induced aging
flavonoid-rich Flemingia macrophylla extract attenuates UVB-induced
in human dermal fibroblasts. Environmental Toxicology. 2020;
skin damage by scavenging reactive oxygen species and inhibiting
MAP kinase and MMP expression. Evid Based Complement Alternat e23065. https://doi.org/10.1002/tox.23065
Med. 2013;2013:696879.