Natural product-based design, synthesis and biological evaluation of 2′,3,4,4′-tetrahydrochalcone analogues as antivitiligo agents

Article abstract of DOI:10.1016/j.bioorg.2019.03.054

A bioactive component, 2′,3,4,4′-tetrahydrochalcone (RY3-a) was first isolated from Vernohia anthelmintica (L.) willd seeds, and a set of its analogs, RY3-a-1–RY3-a-15 and RY3-c were designed and synthesized. Biological activity assays showed that RY3-c exhibited better melanogenesis and antioxidant activity and lower toxicity in comparison with RY3-a and butin. Further study tests showed that RY3-c exhibited better melanogenesis activity compared with the positive control 8-methoxypsoralan (8-MOP) in a vitiligo mouse model, suggesting that RY3-c is a good candidate antivitiligo agent. Mechanistic studies showed that RY3-c could repair cell damage induced by excessive oxidative stress and may exert melanin synthesis activity in the mouse melanoma B16F10 cell line by activating the mitogen-activated protein kinase (MAPK) pathway and the upregulation of c-kit.

Full text of DOI:10.1016/j.bioorg.2019.03.054

BioorganicChemistry87(2019)523–533
Contents lists available at ScienceDirect
Bioorganic Chemistry
journal homepage: www.elsevier.com/locate/bioorg
Natural product-based design, synthesis and biological evaluation of 2′,3,4,
4′-tetrahydrochalcone analogues as antivitiligo agents
⁎
Hui Zhonga, Jia Zhoua, , Xiao-Hong Ana, Ying-Rong Huaa, Yi-Fan Laia, Rui Zhanga,
⁎
⁎
Owais Ahmada, Ye Zhangb,c, , Jing Shanga,
a State Key Laboratory of Natural Medicines, Jiangsu Key Laboratory of TCM Evaluation and Translational Research, School of Traditional Chinese Pharmacy, China
Pharmaceutical University, Nanjing 211198, China
b School of Pharmacy, Guilin Medical University, Guilin 541004, China
c State Key Laboratory for Chemistry and Molecular Engineering of Medicinal Resources, School of Chemistry and Pharmacy, Guangxi Normal University, Guilin 541004,
China
A R T I C L E I N F O
A B S T R A C T
Keywords:
A bioactive component, 2′,3,4,4′-tetrahydrochalcone (RY3-a) was first isolated from Vernohia anthelmintica (L.)
willd seeds, and a set of its analogs, RY3-a-1–RY3-a-15 and RY3-c were designed and synthesized. Biological
activity assays showed that RY3-c exhibited better melanogenesis and antioxidant activity and lower toxicity in
comparison with RY3-a and butin. Further study tests showed that RY3-c exhibited better melanogenesis ac-
tivity compared with the positive control 8-methoxypsoralan (8-MOP) in a vitiligo mouse model, suggesting that
RY3-c is a good candidate antivitiligo agent. Mechanistic studies showed that RY3-c could repair cell damage
induced by excessive oxidative stress and may exert melanin synthesis activity in the mouse melanoma B16F10
cell line by activating the mitogen-activated protein kinase (MAPK) pathway and the upregulation of c-kit.
2′,3,4,4′-Tetrahydrochalcone
Vitiligo
Oxidative stress
MAPK
C-kit
1. Introduction
consequently play an important role in the prevention and treatment of
vitiligo [11]. The potent antivitiligo effect of Gingko biloba [12] inspired
As a disease affecting 1–4% of the world’s population, vitiligo is a
depigmenting disorder of the skin and hair follicles characterized by the
destruction of the melanocytes, mainly exhibiting clinical manifestation
as expanding hypopigmented lesions of the skin [1]. The pathobiology
of vitiligo has been debated for the last six decades, and remains ob-
scure [1,2]. The absence of melanocytes in the skin lesion due to their
destruction is suggested to be the key event in the pathogenesis of vi-
tiligo [3]. Previous work has indicated that oxidative stress could be
considered as one of the important pathogenic events in melanocyte
loss [4,5]. In addition, systemic oxidative stress has been widely re-
ported in vitiligo patients [6–8]. Moreover, a previous study has also
suggested oxidative stress as the initial triggering factor in precipitating
vitiligo [6]. Therefore, oxidative stress is now seriously considered to be
the main pathogenic event in vitiligo [6,9]. Oxidative stress may induce
and mediate multiple genes encoding proteins through several signaling
pathways, including tyrosinase (TYR), microphthalmia-associated
transcription factor (MITF), and MAPK, and ultimately leads to melanin
loss and vitiligo [10]. It is thus widely considered that antioxidant in-
tervention should improve the level of oxidative stress, which may
us to screen for antivitiligo agents from traditional Chinese medicinal
herbs, which are considered to be an important natural biological drug
library and include antioxidant agents. In the search for new antivitiligo
agents with high efficacy and low/no toxicity, considerable attention to
antivitiligo herbs or formulas derived from traditional Chinese medi-
cine herbs is now considered to be a reasonable research option.
Vernohia anthelmintica (L.) willd, which is an annual Chinese med-
icinal herb plant of asteraceae, has been used as a traditional Uyghur
medicine (TUM) for vitiligo treatment since 1763 [13]. As a common
drug for the treatment of vitiligo, Vernohia anthelmintica (L.) willd seeds
exhibit a significant effect in the treatment of melanin deficiency and
vitiligo [14,15]. To identify the active constituents responsible for the
treatment of melanin deficiency and vitiligo in this herb, the extraction,
separation and synthesis of medicinal components from Vernohia an-
thelmintica (L.) willd seeds has been performed. Previous work has in-
dicated that the methanol and chloroform fractions of Vernohia an-
thelmintica (L.) willd seeds exhibit significant antioxidant activities, and
flavonoids may be the active constituents responsible for the treatment
of melanin deficiency and vitiligo [16]. Recently,
a bioactive
⁎ Corresponding authors at: State Key Laboratory of Natural Medicines, Jiangsu Key Laboratory of TCM Evaluation and Translational Research, School of
Traditional Chinese Pharmacy, China Pharmaceutical University, Nanjing 211198, China. School of Pharmacy, Guilin Medical University, Guilin 541004, China.
E-mail addresses: cpuzj_2016@163.com (J. Zhou), zhangye81@126.com (Y. Zhang), shangjing21cn@163.com (J. Shang).
https://doi.org/10.1016/j.bioorg.2019.03.054
Received 2 November 2018; Received in revised form 19 February 2019; Accepted 18 March 2019
Availableonline23March2019
0045-2068/©2019ElsevierInc.Allrightsreserved.

H. Zhong, et al.
BioorganicChemistry87(2019)523–533
Scheme 1. Synthetic route of compound RY3-a. Reagents and conditions: (a) 60% KOH, EtOH.
component butin has been isolated from Vernohia anthelmintica (L.)
willd seeds and was reported to exhibit potent antivitiligo activity in a
mouse model [17]. However, to the best of our knowledge, not all of the
bioactive components have been isolated, and the antivitiligo activities
and the action mechanisms have not been completely investigated.
Herein, the isolation and separation of the active constituents form
Vernohia anthelmintica (L.) willd seeds was thus performed. In the pre-
sent work, the flavonoid compound 2′,3,4,4′-tetrahydrochalcone (RY3-
a) was first isolated from Vernohia anthelmintica (L.) willd seeds. A
biological evaluation assay showed that 2′,3,4,4′-tetrahydrochalcone
exhibited increased antivitiligo activity and no significant toxicity in
comparison with butin. However, due to low yield of 2′,3,4,4′-tetra-
hydrochalcone from the plant material, we accomplished its total
synthesis, which enabled us to synthesize sufficient compounds for
further biological studies. In addition, the significant antivitiligo ac-
tivity, structural simplicity and low toxicity of 2′,3,4,4′-tetra-
hydrochalcone also greatly encouraged us to design and synthesize
additional 2′,3,4,4′-tetrahydrochalcone analogs as antivitiligo agents.
Thus, a set of 2′,3,4,4′-tetrahydrochalcone analogs (RY3-a-1–RY3-a-
15, Scheme 1) were designed and synthesized as antivitiligo agents in
the present work based on the skeleton of 2′,3,4,4′-tetrahydrochalcone.
Unfortunately, preliminary zebrafish model assay results showed that
all of the synthetic 2′,3,4,4′-tetrahydrochalcone analogs, RY3-a-1–RY3-
a-15, exhibited lower melanin synthesis activity than RY3-a, and some
of the analogs (those with the presence of OH near the carbonyl group)
exhibited increased toxicity. Through the study of the structure-activity
relationship of RY3-a, its analogs and butin, the tetrahydro-4-pyrone
ring of butin and the conjugated ketene skeleton of RY3-a analogs in-
spired us to construct a dihydro-4-pyranone ring (Fig. 1) in the RY3-a
moiety and to remove the OH near the carbonyl group. Previous work
has indicated that cyclic flavone derivatives usually exhibited good
antioxidant activity, and it was thus expected that the combination of
tetrahydro-4-pyrone and the conjugated ketene skeleton may lead to
good antioxidant activity and consequently potential antivitiligo ac-
tivity. Thus, a flavone compound (RY-3-c) was finally designed and
synthesized as an antivitiligo agent based on the 2′,3,4,4′-tetra-
hydrochalcone framework (Scheme 2). Antivitiligo activity screening
results showed that in comparison with RY-3-a and RY-3-b, RY-3-c
exhibited increased antioxidant and antivitiligo activities and lower
toxicity. Further tests showed that RY3-c exhibited better melanogen-
esis activity compared with the commercial antivitiligo drug 8-meth-
oxypsoralan (8-MOP) in a vitiligo mouse model, suggesting that RY3-c
should be a good candidate for an antivitiligo agent, which is consistent
with our expectation. In addition, our preliminary mechanistic study
showed that RY3-c could repair cell damage induced by excessive
oxidative stress. Moreover, further investigation into the mechanism of
action revealed that RY-3-c may exert antivitiligo activity via MAPK
signaling, providing good proof-of-concept for our network pharma-
cological prediction studies in the present work.
2. Results and discussion
2.1. Chemistry
2′,3,4,4′-Tetrahydrochalcone, isolated as
a yellow amorphous
powder, showed a molecular ion peak at m/z 271 in its HREI-MS, in-
dicating a molecular formula of C15H12O5 (calcd for [M-H], 271). Its
NMR spectra exhibited the presence of four hydroxyl groups
[δH = 13.62, 10.68, 9.78, 9.12 and δC = 113.35, 148.53, 164.94,
166.10], six aromatic protons [δH = 7.72, 7.33, 6.87, 6.45, 6.34 and
δC = 114.45, 113.35, 115.24, 116.96, 127.07, 131.88], one carbonyl
group [δC = 192.13] and one olefin group [δH = 8.19, 7.26 and
δC = 107.75, 144.69]. These data, together with comparison with the
NMR data of the known 2′,3,4,4′-tetrahydrochalcone, established the
structure of compound RY-3a, named 2′,3,4,4′-tetrahydrochalcone
[18]. The low yields of 2′,3,4,4′-tetrahydrochalcone in Vernohia an-
thelmintica (L.) willd seed prompted us to design, synthesize and ex-
plore
a simple and efficient synthetic route for 2′,3,4,4′-tetra-
hydrochalcone as well as a synthetic method for its analogues. The
synthetic routes are shown in Scheme 1 (compounds RY3-a, RY3-a-
1–RY3-a-15).
As shown in Scheme 1, 2′,3,4,4′-tetrahydrochalcone (RY3-a) and its
derivatives (RY3-a-1–RY3-a-15) were synthesized through a Claisen-
Schmidt reaction [19–21] by the combination of corresponding 2,4-
dihydroxyacetophenone derivatives and protocatechualdehyde deriva-
tives, respectively.
By the analysis of the structure-activity relationship of RY3-a and its
derivatives and butin, a 2′,3,4,4′-tetrahydrochalcone analog RY3-c was
then designed and synthesized as described in Scheme 2. As shown in
Scheme 2, RY3-c-1, which was synthesized by the treatment of 3,4-
dihydroxyl-ethyl benzoate and benzyl bromide in the presence of po-
tassium, was treated with 2,4-dihydroxyl-benzophenone to afford RY3-
c-2. The treatment of RY3-c-2 with Dowex-H+ resin in the presence of
2-propanol provided compound RY3-c-3, which was finally hydrolyzed
to afford RY3-c in the presence of 40% potassium hydrate.
Fig. 1. Design of compound RY-3c.
524

H. Zhong, et al.
BioorganicChemistry87(2019)523–533
Scheme 2. Synthetic route of compound RY3-c. Reagents and conditions: (a) PhBr, K2CO3, CH2Cl2; (b) LiHMDS, tetrahydrofuran (THF), −60 °C; (c) Dowex-H+, 2-
propanol, reflux; (d) Pd-C, H2, MeOH.
The structures of all of the target compounds (RY3-a, RY3-a-
1–RY3-a-15 and RY3-c) were confirmed by 1HNMR, 13CNMR and high
resolution mass spectrometry (HRMS) (Part 4 of the Supplementary
Data).
position (near the carbonyl group) may lead to serious toxicity and thus
should be removed.
Based on the above observation, it should be concluded that this
structural modification of 2′,3,4,4′-tetrahydrochalcone did not improve
the activities of melanotropins, and the presence of a hydroxyl group in
the R1 position may even increase its toxicity. Thus, as described in the
Introduction section, the 2′,3,4,4′-tetrahydrochalcone analog RY3-c
was designed and synthesized to improve the activities of melanotropin
and the toxicity by the combination of RY3-a and butin. The melano-
tropin and tyrosinase enzymatic activities of RY3-c were then assayed
using RY3-a and RY3-b (butin) as the positive controls.
2.2. Evaluation of the antivitiligo and antioxidant activities
2.2.1. Melanogenesis and tyrosinase enzymatic activities in zebrafish
Due to the melanin pigments in their skin, zebrafish have been
considered an efficient whole-animal model for screening melanogenic
inhibitors or stimulators [22]. Changes in the melanin content and
tyrosinase enzymatic activity are the most visual indices to evaluate the
movement of molecules. In general, cutaneous melanin can be first
observed at 24 hpf in the zebrafish retina, and alterations in the tyr-
osinase enzymatic activity can be detected at 21 hpf [23]. To better
identify the changes in melanin content and tyrosinase enzymatic ac-
tivity, the well-known melanogenic inhibitor 1-phenyl-2-thiourea
(PTU) was used to build the pigment model. As shown in Fig. 2A,
compared to the control group and the PTU-2 group, the melanin
content of zebrafish treated with PTU decreased significantly from 6 hpf
to 60 hpf. In the PTU-2 group, zebrafish were treated with PTU from 6
hpf to 35 hpf and then water. The melanin content of zebrafish in this
group was more than that of the PTU-1 group but less than that of the
control group. Fig. 2A also shows that the depigmentation model of
zebrafish, which involved treatment with PTU, was safe and effective.
As shown in Fig. 2A, compounds RY3-a, RY3-a-1, RY3-a-4, RY3-a-7
and RY3-a-10 significantly increased melanin content in zebrafish,
while compounds RY3-a-3, RY3-a-5, RY3-a-6, RY3-a-9, RY3-a-12,
RY3-a-14 and RY3-a-15 could not increase melanin synthesis in zeb-
rafish, indicating that RY3-a, RY3-a-1, RY3-a-4, RY3-a-7 and RY3-a-
10 may be good melanogenic stimulators. Evidently, among all of these
compounds, RY3-a exhibited the best stimulating activity for melano-
synthesis. It was important to note that treatment with RY3-a-3, RY3-
a-5, RY3-a-8 and RY3-a-14 led to the morphological deformity of
zebrafish, indicating that these compounds exhibited obvious toxicity.
This result also implied that the presence of a hydroxyl group in the R1
As shown in Fig. 3A–C, in comparison with the PTU-2 group, the
changes in the melanin content of the entire zebrafish body, and par-
ticularly those of the retina and dorsolateral skin, showed that RY3-a,
RY3-b and RY3-c significantly increased melanin synthesis in zebrafish
in a dose-dependent manner. It was worth noting that RY3-c exhibited
the highest melanotropin activity among these three compounds. The
tyrosinase activities of zebrafish treated with RY3-a, RY3-b and RY3-c
were also measured. It is important to note that, as shown in Fig. 3D,
both RY3-a and RY3-c significantly increased the tyrosinase activity at
concentrations of 20 μM and 40 μM in comparison with the PTU-2
group, while RY3-c exhibited higher tyrosinase activity. In the RY3-b
treatment group, there was no significant difference after RY3-b-
treatment (1, 10, 20, and 40 μM) compared with the PTU-2 group.
Thus, it was obvious that RY3-c exhibited the greatest melanogenic and
tyrosinase enzymatic stimulating activities of these three compounds.
2.2.2. Melanogenesis activity quantified by TYR and MITF
As the rate-limiting enzyme, TYR can convert tyrosine to L-3,4-di-
hydroxyphenylalanine, dopaquinone, dopachrome and subsequently
melanin during the production of melanins [24]. In addition, MITF is
the major transcription factor for the upregulation of the transcription
of melanin genes and the expression of the melanogenic enzyme [25].
Thus, the expression levels of TYR and MITF should be quantified as the
unique detection indices of melanogenesis [26,27]. To evaluate the
melanogenesis activity, B16F10 cells were treated with RY3-a, RY3-b
525

H. Zhong, et al.
BioorganicChemistry87(2019)523–533
Fig. 2. Melanogenesis and tyrosinase enzymatic activities of RY3-a derivatives in the zebrafish model.
and RY3-c for 48 h, and Fontana-Masson staining was performed. As
shown in Fig. 4, all of these compounds increased melanogenesis in
B16F10 cells, while RY3-a and RY3-c exhibited higher melanogenesis
activity than RY3-b. In addition, Fig. 4 also shows that both RY3-a and
RY3-c promoted melanin transfer from the area near the nucleus to the
surrounding membrane area and to the dendrite tips, in particular,
indicating that RY3-a and RY3-c not only promoted melanin synthesis
but also stimulated melanin translocation.
been considered a genetically tractable organism with similarities to
humans. Zebrafish have been used for chemical toxicity testing since
1965 [29]. It is known that 24 hpf is at the end segmentation stage from
embryos to larvae, and the time points of 48 hpf, 72 hpf, 96 hpf, and
120 hpf are specified time points during the developmental progression
of zebrafish. Therefore, the mortality of zebrafish should be recorded at
these time points [30]. Body length and morphology deformity as-
sessments are the major testing indices for toxicity tests; they should be
observed and photographed at the time points specified above. There-
fore, the toxicity of RY3-a was observed in zebrafish, and malforma-
tions were photographed. As shown in Fig. S1-1.A (Supplementary Data
(Part 1)), pericardial edema was observed at the concentration of
100 μM at 96 hpf, while significant malformations were evident at the
concentration of 150 μM from 48 hpf to 120 hpf. As shown in Fig. S1-
1D, in the RY3-b treatment group, pericardial edema was observed at
the concentration of 50 μM at 120 hpf, and significant malformations
were obvious at the concentration of 100 μM from 48 hpf to 120 hpf. It
was worth noting that, as shown in Fig. S1-1G, pericardial edema in the
RY3-c-treated group was observed even at the concentration of 200 μM
at 120 hpf. The body length and morphology deformity assessments
above showed that RY3-c exhibited the lowest toxicity in zebrafish,
while RY3-b exhibited the highest toxicity. As shown in Fig. S1-1B, Fig.
S1-1E, Fig. S1-1H, Fig. S1-1C, Fig. S1-1F, and Fig. S1-1I, taking the
To further evaluate the melanogenesis activity, the in vitro expres-
sion levels of TYR and MITF were then investigated by western blot
assays. As shown in Fig. 5, it was evident that the expression levels of
MITF and TYR remarkably increased in B1F10 cells by treatment with
RY3-c at concentrations of 0.1 μM, 1 μM and 10 μM. However, in the
RY3-a and RY3-b treatment groups, RY3-a and RY3-b only increased
the expression levels of MITF at a concentration of 10 μM. The data
strongly suggested that RY3-c exhibited the best efficacy towards
melanogenesis, which is consistent with the above melanogenesis and
tyrosinase enzymatic activity assay results.
2.2.3. Toxicity tests in the zebrafish model
Previous studies have shown that in comparison to the human re-
ference genome, approximately 70% of human genes have at least one
obvious zebrafish orthologue [28]. Thus, zebrafish (Danio rerio) has
526

H. Zhong, et al.
BioorganicChemistry87(2019)523–533
Fig. 3. Melanogenesis and tyrosinase enzymatic activities of RY3-a, RY3-b and RY3-c in the zebrafish model. (A) Zebrafish were treated with PTU (0.2 mM) from 6
hpf to 35 hpf, and then, PTU was changed to water for another 25 h (from 35 hpf to 60 hpf) in the PTU-2 group. In the PTU-1 group, zebrafish were treated with PTU
(0.2 mM) from 6 hpf to 60 hpf, while water was used in the control group. (B) Zebrafish were treated with PTU (0.2 mM) from 6 hpf to 35 hpf, and then, PTU was
changed to water containing different doses of RY3-a, RY3-b and RY3-c for another 25 h (from 35 hpf to 60 hpf). RY3-a, RY3-b and RY3-c increased melanin
recovery. (C) A homogenate of each group of zebrafish was used to examine melanin content. (D) A homogenate of each group of zebrafish was used to examine the
tyrosinase activity. Data were analyzed by one-way analysis of variance (ANOVA) followed by the post hoc Tukey test. *P < 0.05, **P < 0.01, compared with PTU-
2.
survival and body length of the zebrafish as indices, there were no
significant differences after RY3-c treatment (1, 10, 50, and 100 μM) in
comparison with the control group at 24, 48, 72, 96 and 120 hpf. The
survival rate and the body length of zebrafish were significantly
changed at the concentration of 100 μM in the RY3-a treatment group,
and changes in the RY3-b treatment group also appeared at the con-
centration of 100 μM. The results also confirmed that RY3-c exhibited
the lowest toxicity in zebrafish.
was treated with placebo ointment, and the model group was treated
with 2.5% hydroquinone, which was dispersed well in excipient once
daily for 40 days after hair removal.
As shown in Fig. 6, compared with the control group, the whitening
of the dorsal skin in mice was obviously increased by hydroquinone
treatment. Treatments with 8-MOP (8.5 μg/day) and RY3-c (10 μg/day)
were performed for 40 days, respectively. The hair color of mice in the
8-MOP treatment group and RY3-c treatment group turned sig-
nificantly darker compared with the model group. As shown in Fig. S1-2
(Supplementary Data (Part 1)), the effects of RY3-c on the numbers of
melanin-containing epidermal cells, melanin-containing follicles and
the basal melanocytes in the shaved skin areas of the mice were ob-
served by hematoxylin and eosin staining. The model group showed a
significant decrease in melanin-containing follicles compared with the
control group, and the number of melanin-containing hair follicles in
the treatment areas were significantly increased for the 8-MOP and
RY3-c treatment groups, indicating high melanogenic activity of RY3-
c. Obviously, RY3-c even exhibited higher melanogenic activity than 8-
MOP, indicating that RY3-c is a good modulator of melanogenesis. This
result was highly consistent with those of the zebrafish model and the
B16F10 cell assays.
In brief, the high melanogenic activity and the low toxicity of RY3-c
rendered it a good candidate as an antivitiligo agent. Thus, RY3-c was
selected for further investigation in a mouse model.
2.2.4. Effect of RY3-c on melanin synthesis in the vitiligo mouse model
induced by hydroquinone
The mouse model of vitiligo induced by hydroquinone has been
widely used for screening antivitiligo agents [31,17]. Forty-eight male
C57BL/6 mice were randomly divided into four groups. Ointments in-
cluding hydroquinone, the commercial antivitiligo drug 8-MOP and
RY3-c were studied, and the formulations of the ointments are de-
scribed in the Supplementary Data (Part 2). Hair was removed via
shaving (1.5 × 1.5 cm) the dorsal skin of the mice. The control group
527

H. Zhong, et al.
BioorganicChemistry87(2019)523–533
Fig. 4. The melanogenic activities of RY3-a, RY3-b and RY3-c in the B16F10 cell model. B16F10 cells were stained with Masson–Fontana ammoniacal silver stain.
B16F10 cells were plated on 13 mm glass coverslips in a six-well plate and then treated with different concentrations of RY3-a, RY3-b and RY3-c for 48 h.
2.3. Investigation of the mechanism of action
green fluorescence between the control group and the H2O2 model
group was remarkable. Compared with the model group, the ROS
content was decreased significantly and concentration-dependently by
RY3-c treatment at concentrations of 1 μM, 10 μM, 20 μM, and 40 μM,
while RY3-b treatment exhibited no obvious change in ROS content.
This result indicated that RY3-c could reverse the ROS content of
zebrafish induced by H2O2 and could thus exhibit potent antioxidant
activity. It was interesting to point out that the fluorescence intensities
in the RY3-a treatment groups (1 μM, 10 μM, 20 μM, and 40 μM) were
clearly lower than that in the H2O2-model group, and the ROS content
decreased to its lowest level at a concentration of 10 μM while it was
increased at concentrations of 20 μM and 40 μM. This result implied
2.3.1. Repair of damage induced by oxidative stress
Antioxidant activity in zebrafish and B16F10 cells. Experimental and
clinical evidence suggests that oxidative stress plays a vital role in the
pathogenesis of the depigmentation process, and high epidermal levels
of reactive oxygen species (ROS) have been discovered in vitiligo skin
[32]. To investigate whether RY3-a, RY3-b and RY3-c exert melanin
synthesis activity through antioxidant activity, the ROS fluorescent
probe 2,7-dichlorodihydrofluorescein diacetate (DCFH-DA) and the
superoxide-specific fluorescent probe dihydroethidium (DHE) were
used in zebrafish and B16F10 cells. As shown in Fig. 7, the difference in
Fig. 5. Effects of RY3-a, RY3-b and RY3-c on the expression levels of TYR and MITF in the B16F10 cell model. The expression levels of TYR and MITF were examined
in B16F10 cells by treatment with RY3-a, RY3-b and RY3-c for 48 h. The results were normalized to β-actin expression.
528

H. Zhong, et al.
BioorganicChemistry87(2019)523–533
Fig. 6. Effect of compound RY3-c on hair growth and hair pigmentation in the vitiligo mouse model induced by hydroquinone. Photographs were taken to document
the gross appearance of control and stressed mice on day 0 and day 12, respectively.
that RY3-a might induce permanent cell injury or mitochondrial im-
pairment and thus increased ROS production in injured cells at high
concentrations of RY3-a [33]. This phenomenon indicated that RY3-a
could exhibit potent antioxidant activity at a low concentration and
induce apoptosis and increase ROS production at high concentrations,
which is similar to the effects of the flavonoid derivatives galangin,
luteolin and myricetin [34]. To illustrate the antioxidant activity of
these three compounds, the superoxide-specific fluorescent probe di-
hydroethidium (DHE) was used in B16F10 cells. Generally, the dim-
ming of red fluorescence indicates good antioxidant activity in com-
parison with the model and positive control. Images of DHE-stained
B16F10 cells are shown in Fig. S1-3 (Supplementary Data (Part 1)). Fig.
S1-3 show that the fluorescence intensity of the model group (H2O2-
treated) was higher than that of the control group, while RY3-a and
RY3-c decreased the superoxide anions detected by lower DHE fluor-
escence compared to the model group and the vitamin C treatment
group. RY3-b showed no obvious anti-oxygenation. Obviously, the
order of antioxidant activity of these three compounds should be listed
as follows: RY3-c > RY3-a > vitamin C > RY3-b.
suggested to be the initial pathogenic event in melanocyte degeneration
and early apoptosis [35]. The cytoskeletal damage and apoptosis of
H2O2-treated B16F10 cells were measured by phalloidin staining and a
flow cytometry assay, respectively. As shown in Fig. 8, H2O2 treatment
markedly induced morphologic changes in B16F10 cells compared with
the control group. Fig. 8 also demonstrated that the morphology of
B16F10 cells showed no obvious change after H2O2 treatment for 4 h,
while dendrite tips appeared after treatment with RY3-a and RY3-c at a
concentration of 10 μM. In contrast, cytomorphological alterations of
B16F10 cells have been observed by RY3-b treatment under the same
conditions. As shown in Fig. S1-4 (Supplementary Data (Part 1)), the
apoptosis of H2O2-treated B16F10 cells was measured by a flow cyto-
metry assay; the flow cytometry assay showed that the apoptotic rate
evidently increased from 7.46% to 13.9% after H2O2 treatment for 4 h
compared to the control group. It is worth noting that pretreatment
with RY3-a and RY3-c could inhibit H2O2-induced apoptosis, and the
apoptotic rates were reduced to 8.30% and 6.30% compared to the
H2O2-treatment group (13.9%), respectively, indicating that RY3-a and
RY3-c could repair the cell damage induced by excessive oxidative
stress. Clearly, as shown in Fig. 8, RY3-c showed better repair ability
than RY3-a.
Inhibition of H2O2-induced cytoskeleton damage and apoptosis in
B16F10 cells. In vitiligo epidermis, H2O2 accumulation has been
Fig. 7. RY3-a, RY3-b and RY3-c affected the levels of ROS in the zebrafish model. Zebrafish (7 dpf) were treated with different doses of RY3-a, RY3-b and RY3-c for
24 h and then H2O2 (500 µM) for 4 h. DCFH-DA (10 µM) was added to zebrafish for 30 min, and then images were captured.
529

H. Zhong, et al.
BioorganicChemistry87(2019)523–533
Fig. 8. Effects of RY3-a, RY3-b and RY3-c on H2O2-induced cytoskeleton damage of B16F10 cells. B16F10 cells were treated with different doses of RY3-a, RY3-b
and RY3-c for 48 h and then H2O2 (100 µM) for 4 h. Cells were incubated with 10 µM phalloidin for 30 min. Scale bars, 100 µm.
2.3.2. MAPK signaling pathway of RY3-c
and some its analogs CY3-a-1–CY3-a-15 and CY3-c were designed and
synthesized as antivitiligo agents. We identified compound CY3-c,
which exhibited higher melanogenic and antioxidant activities and
lower toxicity in comparison with CY3-a and butin (CY3-b). The viti-
ligo mouse model study showed that CY3-c exhibited better melano-
genesis activity compared with the commercial antivitiligo drug 8-
MOP. Mechanistic studies showed that RY3-c exhibited potent anti-
oxidant activity and could repair cell damage induced by excessive
oxidative stress. Further investigation indicated that the p38-MAPK and
ERK1/2 pathway is the key pathway affected by RY3-c during melanin
synthesis and that c-kit is the key regulatory protein, indicating that the
MAPK signaling pathway exhibits a high degree of correlation and thus
may play an important role in melanin synthesis. This work may pro-
vide important information for understanding the antivitiligo me-
chanism of RY3-c and the rational design of antioxidant-based anti-
vitiligo agents.
Previous studies have indicated that oxidative stress could mediate
the signaling pathways involved in melanin synthesis and vitiligo, in-
cluding TYR, MITF, and MAPK [10]. To identify the involved signaling
pathways, the 3D structure of RY3-c was submitted to the versatile web
server ChemMapper to study the pharmacology and chemical structure
associations [6]. The top 100 related potential targets obtained from
ChemMapper were sent to DABID bioinformatics resources, all of the
targets interacting with RY3-c were mapped onto the KEGG pathways,
and cytoscape V3.2.0 software was employed to establish the network
(Fig. S3-1 in the Supplementary Data (Part 3)). The network pharma-
cology research results showed that the MAPK signaling pathway ex-
hibited a high degree of correlation and thus may play an important
role in melanin synthesis (Fig. S3-2). Fig. S3-2 shows that c-kit played a
key role in melanogenesis via the MAPK signaling pathway and was
thus considered to be the target protein for melanin synthesis.
To confirm this signaling pathway, western blot assays were then
carried out, and the results are shown in Fig. 9. As shown in Fig. 9A, the
expression levels of c-Kit increased in a dose-dependent manner by
treatment with RY3-c at concentrations of 0.1, 1 and 10 μM. To better
understand whether the expression of c-Kit is involved in melanin
synthesis, the disruption of c-Kit expression was performed using
siRNA. Optimized transfection conditions of B16F10 cells were de-
termined as shown in Fig. 9B. It was found that the combination of
70 μM siRNA and 12.5 μL of Lipo2000 exhibited a better outcome and
was thus used for subsequent experiments. As shown in Fig. 9C, the
expression of TYR and MITF was reduced significantly by combined
siRNA and RY3-c treatment, which is consistent with the result ob-
tained by treating the cells with RY3-c alone. It is known that c-Kit
plays a key role in melanogenesis. To further clarify the mechanism
underlying the effect of RY3-c on melanogenesis, the MAPK in-
tracellular signal transduction cascade was examined. As shown in
Fig. 9D, 10 μM RY3-c induced p38 and p-JNK phosphorylation, while
the t-JNK phosphorylation remained unchanged. p38 phosphorylation
increased from 5 min to 120 min, while ERK1/2 phosphorylation was
induced at 15 min and reached a peak at 30 min, indicating that RY3-c
promoted the p38-MAPK and ERK1/2 pathway. To gain further support
for this hypothesis, the ERK inhibitor PD98059 and the p38 inhibitor
SB203580 were then used. As shown in Fig. 9E and F, western blotting
analysis showed that the expression of TYR and MITF was significantly
reduced by treatment with the ERK inhibitor PD98059 and the p38
inhibitor SB203580. This result indicated that the p38-MAPK and
ERK1/2 pathway is the key regulatory pathway affected by RY3-c
during melanin synthesis and that c-kit is the key regulatory protein,
which is consistent with the above network pharmacological prediction
results.
4. Experimental procedures
4.1. Materials
All of the chemical reagents and solvents used were of analytical grade.
The following reagents were purchased from J&K Scientific, Ltd.
(Shanghai, China): butin, 2,4-dihydroxyacetophenone, 2-dihydrox-
yacetophenone, 4-dihydroxyacetophenone, 3,4-dihydroxy-benzaldehyde,
3-dihydroxy-benzaldehyde, and 4-dihydroxy-benzaldehyde. MR spectra
were recorded on a Bruker AV-500 NMR spectrometer. Mass spectra were
determined on a Q-TOF ESI spectrometer. All compounds were de-
termined by a PE2400 Type II elemental analyzer (U.S. PE firm).
4.2. Chemistry
4.2.1. Isolation procedure for compound RY3-a
RY3-a was isolated from the seeds of Vernohia anthelmintica (L.)
willd. The seeds were washed with water and then soaked in petroleum
ether for 24 h. The seeds were extracted by water, and the crude extract
was extracted by diethyl ether and ethyl acetate successively. The ethyl
acetate-extracted fraction was separated by octadecylsilyl, and a yellow
solid was acquired. Yield, 0.081%; 1HNMR (500 MHz, DMSO‑d6): δ
6.34 (s, 1H, H), 6.46 (d, 1H, J = 20 Hz, H), 6.87 (d, 1H, J = 10 Hz, H),
7.26 (d, 1H, J = 20 Hz, H), 7.33 (s, 1H), 7.72 (d, 2H, J = 10 Hz, H),
8.19 (d, 1H, J = 15 Hz, H), 9.12 (s, 1H), 9.78 (s, 1H), 10.68 (s, 1H),
13.62 (s, 1H). 13CNMR (126 MHz, DMSO‑d6): δ102.45, 107.75, 113.35,
114.45, 115.24, 116.96, 122.20, 127.07, 131.88, 144.69, 145.45,
148.53, 164.94, 166.10, 192.13. Q-TOF (m/z): calculated for C15H12O5
[M-H]−: 271.0760, found: 271.0749.
3. Conclusion
4.2.2. General synthetic procedure for compounds RY3-a (RY3-a-1-RY3-a-
15)
In summary, the bioactive component 2′,3,4,4′-tetrahydrochalcone
2,4-Dihydroxyacetophenone derivatives (44.7 mmol) in EtOH
(20 mL) were added to aqueous potassium hydroxide (41.6 mL, 60% w/
(CY3-a) was first isolated from Vernohia anthelmintica (L.) willd seeds,
530

H. Zhong, et al.
BioorganicChemistry87(2019)523–533
Fig. 9. Effect of compound RY3-c on the expression levels of relevant proteins in B16F10 cells. (A) Effect of compound RY3-c on the expression of c-kit. (B) Inhibition
of c-kit expression using siRNA transfected into B16F10 cells. (C) Effect of compound RY3-c on the expression of MITF and TYR after c-kit siRNA treatment in B16F10
cells. (D) Effect of compound RY3-c on the expression of p38, ERK, and JNK and the phosphorylation of p38, ERK, and JNK. (E) Effect of compound RY3-c on the
expression of MITF and TYR after PD98059 treatment in B16F10 cells. (F) Effect of compound RY3-c on the expression of MITF and TYR after SD203580 treatment in
B16F10 cells.
w) under magnetic stirring for 10 min at room temperature followed by
the addition of 2,3-dihydroxybenzaldehyde derivatives (45.9 mmol).
The mixture was stirred magnetically at 85 °C for 2 h and then poured
onto ice. The reaction mixture was acidified to a pH = 4 using cold
hydrochloric acid and extracted with EtOAc. The combined EtOAc ex-
tracts were washed with brine solution, dried with Na2SO4 and con-
centrated under reduced pressure. Purification of the crude product by
flash chromatography (20% EtOAc/hexane) yielded RY3-a-1–RY3-a-
15.
1H, J = 15 Hz), 7.92 (dd, 2H, J = 10 Hz), 8.00 (d, 1H, J = 15 Hz), 8.22
(d, 1H, J = 10 Hz), 10.74 (s, 1H), 13.41 (s, 1H);13CNMR (126 MHz,
DMSO‑d6): δ 103.12, 108.84, 113.50, 121.83, 129.41, 129.48, 131.15,
133.65, 135.13, 144.10, 165.91, 166.32, 191.94; Q-TOF (m/z): calcu-
lated m/z for C15H12O3 [M-H]−: 239.0780, found: 239.0718.
RY3-a-4: Yield, 77.6%; 1HNMR (500 MHz, DMSO‑d6): δ 6.81 (m,
1H, J = 15 Hz), 6.90 (s, 2H), 7.23 (m, 1H, J = 5 Hz), 7.57 (m, 2H,
J = 50 Hz), 8.02 (d, 2H, J = 10 Hz), 9.03 (s, 1H), 9.64 (s, 1H), 10.34 (s,
1H); 13CNMR (126 MHz, DMSO‑d6): δ115.77, 115.88, 116.22, 118.94,
122.28, 126.96, 129.98, 131.34, 144.08, 146.04, 148.86, 162.32,
187.52; Q-TOF (m/z): calculated m/z for C15H12O4 [M-H]−: 255.0770,
found: 255.0687.
RY3-a-1: Yield, 73.2%; 1HNMR (500 MHz, DMSO‑d6): δ 6.30 (m,
1H, J = 10 Hz), 6.43 (dd, 1H, J = 10 Hz), 6.85 (d, 2H, J = 10 Hz), 7.75
(d, 4H, J = 10 Hz), 8.16 (dd, 1H, J = 10 Hz), 10.14 (s, 1H), 10.68 (s,
1H), 13.62 (s, 1H). 13CNMR (126 MHz, DMSO‑d6): δ 103.10, 108.58,
113.51, 116.33, 117.90, 126.24, 133.28, 131.66, 160.74, 165.42,
RY3-a-5: Yield, 81.2%;1HNMR (500 MHz, DMSO‑d6): δ 6.84 (d, 1H,
J = 10 Hz), 6.99 (m, 2H, J = 15 Hz), 7.24 (d, 1H, J = 10 Hz), 7.32 (s,
1H), 7.55 (t, 1H, J = 20 Hz), 7.74 (s, 1H), 8.23 (d, 1H, J = 20 Hz);
13CNMR (126 MHz, DMSO‑d6): δ 116.28, 117.95, 118.17, 119.51,
121.15, 123.39, 126.40, 131.04, 136.43, 146.35, 146.63, 162.47,
193.85; Q-TOF (m/z): calculated m/z for C15 H12O 4 [M-H]−: 255.0770,
found: 255.0603.
166.28, 192.01; Q-TOF (m/z): calculated m/z for C15H12O6 [M-H]−
:
255.0770, found: 255.0674.
RY3-a-2: Yield, 65.2%; 1HNMR (500 MHz, DMSO‑d6): δ 6.30 (d, 1H,
J = 5 Hz), 6.42 (m, 1H, J = 15 Hz), 6.88 (m, 1H, J = 10 Hz), 7.26 (m,
2H, J = 20 Hz), 7.32 (d, 1H, J = 10 Hz), 7.70 (d, 1H, J = 20 Hz), 7.88
(d, 1H, J = 25 Hz), 8.18 (d, 1H, J = 15 Hz), 9.63 (s, 1H), 10.75 (s, 1H),
13.38 (s, 1H).; 13CNMR (126 MHz, DMSO‑d6): δ103.08, 108.77, 113.54,
115.92, 118.35, 120.44, 121.63, 130.35, 133.63, 136.39, 144.38,
158.21, 165.42, 166.25, 191.95; Q-TOF (m/z): calculated m/z for
RY3-a-6: Yield, 53.2%; 1HNMR (500 MHz, DMSO‑d6): δ 6.84 (d, 1H,
J = 5 Hz), 7.19 (dd, 1H, J = 5 Hz), 7.28 (s, 1H,), 7.32 (s, 1H), 7.57 (t,
1H, J = 15 Hz), 7.61 (s, 1H), 7.66 (t, 1H, J = 15 Hz), 8.11 (m, 2H,
J = 10 Hz);13CNMR (126 MHz, DMSO‑d6):δ 115.97, 116.32, 118.76,
122.76, 126.54, 128.76, 129.19, 133.20, 138.57, 145.53, 146.21,
149.59; Q-TOF (m/z):calculated m/z for C15H12O3 [M-H]−: 239.0780,
found: 239.1026.
C
15H12O
[M-H]−: 255.0770, found: 255.0675.
6
RY3-a-3: Yield, 80.1%;1HNMR (500 MHz, DMSO‑d6): δ 6.34 (d, 1H,
J = 5 Hz), 6.46 (dd, 1H, J = 10 Hz), 7.49 (dd, 3H, J = 5 Hz), 7.82 (d,
531

H. Zhong, et al.
BioorganicChemistry87(2019)523–533
RY3-a-7: Yield, 84.2%; 1HNMR (500 MHz, DMSO‑d6): δ 6.87 (m,
of 3,4-dihydroxybenzoic acid ethyl ester (1.82 g, 10 mmol) were pro-
tected by benzyl (RY3-c) and dissolved in dry THF (20 mL). The mix-
ture was then cooled to −60 °C and to 4.36 g of RY3-c, and dry THF
(40 mL) was added dropwise. This mixture was stirred for another
45 min at −60 °C and then incubated at room temperature overnight.
The reaction mixture was poured on ice and acidified with 2.0 M HCl.
The THF was removed, the crude 1-(3,4-bis-benzyloxy-phenyl)-3-(2,4-
dihydroxy-phenyl)-propane-1,3-dione was dissolved in 2-propanol
(50 mL), 0.5 g Dowex W50 × 8 (H+ form) was added, and the reaction
was heated under reflux for 24 h under an atmosphere of dry nitrogen.
The solids were filtered, and then the reaction mixture was poured on
water. The solids were filtered and dissolved in MeOH (50 mL), and Pd-
C (1 g) was added under an atmosphere of dry hydrogen at room
temperature overnight. The solids were filtered and MeOH was re-
moved. Purification of the crude product by flash chromatography
(30% EtOAc/hexane) yielded RY3-c (1.28 g, 32.4%) as a white solid.
RY3-c: 1HNMR (500 MHz, DMSO‑d6): δ 6.62 (s, 1H), 6.92 (m, 2H,
J = 10 Hz), 6.95 (d, 1H, J = 5 Hz), 7.40 (dd, 2H, J = 10 Hz), 7.88 (d,
1H, J = 10 Hz), 9.61 (s, 2H), 10.77 (s, 1H). 13CNMR (126 MHz,
DMSO‑d6): δ102.83, 104.97, 113.66, 115.24, 116.47, 118.99, 122.62,
126.94, 146.16, 149.61, 157.81, 163.02, 163.07, 176.63. Q-TOF (m/z):
calculated m/z for C15H10O5 [M-H]−: 269.0500, found: 269.0536.
4H, J = 40 Hz), 7.68 (dd, 4H, J = 50 Hz), 8.04 (dd, 2H, J = 15 Hz),
10.20 (s, 1H); 13CNMR (126 MHz, DMSO‑d6): δ 115.78, 116.27, 119.04,
126.47, 129.95, 131.18, 131.40, 143.65, 160.30, 162.38, 187.57. Q-
TOF (m/z): calculated m/z for C15H12O3 [M-H]−: 239.0780, found:
239.0817.
RY3-a-8: Yield, 77.6%;1HNMR (500 MHz, DMSO‑d6): δ 6.86 (d, 2H,
J = 10 Hz), 7.00 (t, 2H, J = 5 Hz), 7.55 (m, 1H, J = 25 Hz), 7.78 (m,
2H, J = 25 Hz), 7.83 (d, 2H, J = 15 Hz), 8.25 (d, 1H, J = 5 Hz), 10.22
(s, 1H,), 12.79 (s, 1H); 13CNMR (126 MHz, DMSO‑d6): δ116.40, 118.20,
119.53, 121.05, 126.06, 131.12, 131.99, 136.57, 146.13, 161.11,
162.54, 194.04.; Q-TOF (m/z): calculated m/z for C15 H12O3 [M-H]−
:
239.0780, found: 239.0942.
RY3-a-9: Yield, 71.3%;1HNMR (500 MHz, DMSO‑d6): δ 6.86 (d, 2H,
J = 10 Hz), 7.57 (t, 2H, J = 20 Hz), 7.66 (t, 1H, J = 10 Hz), 7.74 (m,
4H, J = 15 Hz), 8.12 (d, 2H, J = 5 Hz), 10.12 (s, 1H,); 13CNMR
(500 MHz, DMSO‑d6): δ116.32, 119.01, 126.26, 128.80, 129.18,
131.49, 133.26, 138.47, 145.00, 160.65; Q-TOF (m/z): calculated m/z
for C15H12O2 [M-H]−: 223.0890, found: 223.0990.
RY3-a-10: Yield, 64.2%;1HNMR (500 MHz, DMSO‑d6): δ 6.88 (ddd,
1H, J = 20 Hz), 6.92 (m, 2H, J = 5 Hz), 7.21 (m, 1H, J = 5 Hz), 7.28
(m, 2H, J = 20 Hz), 7.60 (d, 1H, J = 15 Hz), 7.81 (d, 1H, J = 15 Hz),
8.07 (m, 2H, J = 30 Hz), 9.60 (s, 1H), 10.41 (s, 1H); 13CNMR
(500 MHz, DMSO‑d6): δ 115.60, 115.87, 117.98, 120.13, 122.45,
129.62, 130.33, 131.63, 136.66, 143.43, 158.20, 162.65, 187.65; Q-
TOF (m/z): calculated m/z for C15H12O3[M-H]−: 239.0780, found:
239.0740.
4.3. Biological assays
The procedures for the biological assays are described in the
Supplementary Data (Part 1–Part 3). The materials, instrumentation
and methods for the cytotoxicity assay, apoptosis analysis and western
blot assays were reported previously [36–38]. Animal handling proce-
dures conformed to the guidelines of the animal ethics committee of the
Chinese Ministry of Health and the animal experiment standards, and
they were approved by the animal management committee of China
Pharmaceutical University.
RY3-a-11: Yield, 60.3%;1HNMR (500 MHz, DMSO‑d6): δ 6.86 (d,
2H, J = 10 Hz), 6.91 (d, 2H, J = 10 Hz), 7.65 (s, 1H,), 7.69 (s, 1H), 7.72
(d, 2H, J = 5 Hz), 8.05 (d, 2H, J = 5 Hz), 10.01(s, 1H), 10.32 (s, 1H);
13CNMR (126 MHz, DMSO‑d6):115.79, 116.27, 119.04, 126.48, 129.96,
131.29, 143.65, 160.30, 162.38, 187.58; Q-TOF (m/z): calculated m/z
for C15H12O3 [M-H]−: 239.0780, found: 239.0907.
RY3-a-12: Yield, 52.3%; 1HNMR (500 MHz, DMSO‑d6) δ 6.90 (m,
1H, J = 10 Hz), 7.26 (m, 2H, J = 30 Hz), 7.33 (d, 1H, J = 10 Hz), 7.59
(t, 2H, J = 15 Hz), 7.68 (m, 2H, J = 25 Hz), 7.84 (d, 1H, J = 15 Hz),
8.15 (m, 2H, J = 5 Hz), 9.64(s, 1H). 13CNMR (126 MHz, DMSO‑d6) δ
115.76, 118.33, 120.36, 122.44, 128.98, 129.28, 130.38, 133.59,
136.42, 138.10, 144.76, 158.23, 189.76; Q-TOF (m/z):calculated m/z
for C15H12O2[M-H]−: 223.0890, found: 223.0866.
Disclosure of potential conflicts of interest
The authors declare no competing financial interests.
Acknowledgments
RY3-a-13: Yield, 55.6%;1HNMR (500 MHz, DMSO‑d6): δ 6.93 (d,
2H, J = 10 Hz), 7.46 (m, 3H, J = 20 Hz), 7.70 (d, 1H, J = 15 Hz), 7.88
(dd, 2H, J = 10 Hz), 7.92 (dd, 1H, J = 15 Hz), 8.11 (s,2H), 10.43(s,
1H);13CNMR (126 MHz, DMSO‑d6); 13CNMR (126 MHz, DMSO‑d6): δ
115.89, 122.66, 129.16, 129.37, 129.62, 130.78, 131.67, 135.39,
143.18, 162.70, 187.66. Q-TOF (m/z): calculated m/z for C15H12O2[M-
H]−: 223.0890, found: 223.0802.
This study was supported by the National Natural Science
Foundation of China (No. 81874331), the Postgraduate Research &
Practice Innovation Program of Jiangsu Province (KYCX17_0669) and
the Open Project for Chemistry and Molecular Engineering of Medicinal
Resources (Guangxi Normal University) (CMEMR2017-B02).
Appendix A. Supplementary material
RY3-a-14: Yield, 63.6%; 1HNMR (500 MHz, DMSO‑d6): δ 7.03 (dt,
2H, J = 20 Hz), 7.49 (m, 3H, J = 10 Hz), 7.59 (t, 1H, J = 10 Hz), 7.86
(d, 2H, J = 15 Hz), 8.06 (d, 1H, J = 20 Hz),8.27 (dd,1H,
J = 20 Hz),12.53 (s,1H); 13CNMR (126 MHz, DMSO‑d6): δ 118.21,
119.65, 122.29, 129.53, 131.36, 134.93, 136.79, 145.25, 162.36,
194.11; Q-TOF (m/z): calculated m/z for C15H12O2 [M-H]−:223.0890,
found: 223.0850.
Supplementary data to this article can be found online at https://
doi.org/10.1016/j.bioorg.2019.03.054.
References
[1] E. Nicolaidou, A.D. Katsambas, Pigmentation disorders: hyperpigmentation and
hypo-pigmentation, Clin. Dermatol. 32 (2014) 66–72.
RY3-a-15: Yield, 70.3%; 1HNMR(500 MHz, DMSO‑d6):δ 7.49 (m,
3H, J = 5 Hz), 7.60 (t, 2H, J = 15 Hz), 7.70 (t, 1H, J = 15 Hz), 7.78 (t,
1H, J = 15 Hz), 7.92 (m, 2H, J = 5 Hz), 7.96 (d, 1H, J = 20 Hz), 8.17
(m,2H, J = 10 Hz);13CNMR (126 MHz, DMSO‑d6): δ 122.62, 129.01,
129.29, 129.38, 129.41, 131.13, 133.63, 144.52, 189.74; Q-TOF (m/z):
calculated m/z for C15H12O [M+Na]+: 231.0830, found: 231.0815.
[2] N. Malhotra, M. Dytoc, The pathogenesis of vitiligo, J. Cutan. Med. Surg. 17 (2013)
153–172.
[3] I.C.L. Poole, P.K. Das, R.M.J.G.J. van den Wijngaard, J.D. Bos, W. Westerhof,
Review of the etiopathomechanism of vitiligo: a convergence theory, Exp.
Dermatol. 2 (1993), 145–153.
[4] K.U. Schallreuter, Successful treatment of oxidative stress in vitiligo, Skin
Pharmacol. Phys. 12 (1999) 132–138.
[5] V. Maresca, M. Roccella, F. Roccella, E. Camera, G. Del Porto, S. Passi,
P. Grammatico, M. Picardo, Increased sensitivity to peroxidative agents as a pos-
sible pathogenic factor of melanocyte damage in vitiligo, J. Invest. Dermatol. 109
(1997) 310–313.
4.2.3. Synthesis of RY3-c
2,4-Dihydroxyacetophenone was dissolved in dry THF, and the
mixture was cooled to −60 °C under an atmosphere of dry nitrogen.
LiHMDS (1.6 M in hexane, 30 mmol) was added dropwise, and the re-
action was stirred for 45 min at −30 °C. The phenolic hydroxyl groups
[6] N.C. Laddha, M. Dwivedi, M.S. Mansuri, A.R. Gani, M. Ansarullah,
A.V. Ramachandran, S. Dalai, R. Bequm, Vitiligo: interplay between oxidative stress
and immune system, Exp. Dermatol. 22 (2013) 245–250.
532

H. Zhong, et al.
BioorganicChemistry87(2019)523–533
[7] A.A. Shah, A.A. Sinha, Oxidative stress and autoimmune skin disease, Eur. J.
Dermatol. 23 (2013) 5–13.
[25] J.H. Ha, Y.J. Jeong, S.H. Xuan, J.Y. Lee, J. Park, S.N. Park, Methyl-2-acetylamino-3-
(4-hydroxyl-3,5-dimethoxybenzoylthio) propanoate suppresses melanogenesis
through ERK signaling pathway mediated MITF proteasomal degradation, J.
Dermatol. Sci. 91 (2018) 142–152.
[8] L. Guerra, E. Dellambra, S. Brescia, D. Raskovic, Vitiligo: pathogenetic hypotheses
and targets for current therapies, Curr. Drug. Metab. 11 (2010) 451–467.
[9] S.J. Glassman, S.B. Heidelberg, ROS and Vitiligo, 2014, 3677.
[10] A. Slominski, D.J. Tobin, S. Shibahara, J. Wortsman, Melanin pigmentation in
mammalian skin and its hormonal regulation, Physiolo. Rev. 84 (2004) 1155–1228.
[11] G. Akoglu, S. Emre, A. Metin, A. Akbas, A. Yorulmaz, S. Isikouqlu, S. Sener,
F. Kilinc, Evaluation of total oxidant and antioxidant status in localized and gen-
eralized vitiligo, Clin. Exp. Dermatol. 38 (2013) 701–706.
[26] R. Zhang, R. Fan, Z. Cheng, X. Tian, J. Liu, Regulation of TYR and MITF mRNA
expression by CDK5 in alpaca melanocytes, Chinese J. Anim. Veter. Sci. 42 (2011)
1712–1717.
[27] Y. Wakabayashi, H. Nakajima, G. Imokawa, Abrogating effect of N-linked carbo-
hydrate modifiers on the stem cell factor and endothelin-1-stimulated epidermal
pigmentation in human epidermal equivalents, J. Dermatol. Sci. 69 (2013)
215–228.
[12] O. Szczurko, N. Shear, A. Taddio, H. Boon, Ginkgo biloba for the treatment of vitilgo
vulgaris: an open label pilot clinical trial, BMC. Complem. Altern. M. 11 (2011)
21–29.
[28] F. Zhang, W. Qin, J.P. Zhang, C.Q. Hu, Antibiotic toxicity and absorption in zeb-
rafish using liquid chromatography-tandem mass spectrometry, Plos One 10 (2015)
e0124805.
[13] Y.S. Tuerxun Drug Standard of Ministry of Public Health of the People’s Republic of
China (Uygur Medicine), Urumqi Science and Technology Publishing House:
Xinjiang. 1999, 150–153.
[29] N.A. Ducharme, D.M. Reif, J.A. Gustafsson, M. Bondesson, Reprod Toxicol,
Comparison of toxicity values across zebrafish early life stages and mammalian
studies: implications for chemical testing, Reprod. Toxicol. 55 (2015) 3–10.
[30] W.H. Peng, Y.C. Lee, Y.P. Chau, K.S. Lu, H.N. Kung, Short-term exposure of zeb-
rafish embryos to arecoline leads to retarded growth, motor impairment, and somite
muscle fiber changes, Zebrafish 12 (2015) 58–70.
[14] D.R. Shi, L.S. Ha, X. Ru, H.F. Gan, J.L. Liu, X.M. Pu, Clinical efficacy of treatment of
Vernonia anthelmintica (L.) Willd on vitiligo, Xinjiang Med. J. 33 (2003) 51–52.
[15] Z.Q. Ma, H. Hu, T.T. He, H. Guo, M.Y. Zhang, M.W. Chen, Y.T. Wang, An assessment
of traditional Uighur medicine in current Xinjiang region (China), Afr. J. Tradit.
Complem. 11 (2014) 301–314.
[31] S.X. Huo, X.M. Liu, C.H. Ge, L. Gao, X.M. Peng, P.P. Zhao, M. Yan, The effects of
galangin on a mouse model of vitiligo induced by hydroquinone, Phytother. Res. 28
(2015) 1533–1538.
[16] A.K. Patel, S.A. Bhalerao. Ethnobotany, phytochemistry and pharmacological eva-
luation of vernonia anthelmintica (l.) willd.: an overview, Asian J. Sci. Technol.
8(2017) 6258–6262.
[32] R. Colucci, F. Dragoni, S. Moretti, Oxidative stress and immune system in vitiligo
and thyroid diseases, Oxid. Med. Cell. Longev. (2015) 631927.
[33] M.L. Dell'Anna, V. Maresca, S. Briganti, E. Camera, M. Falchi, M. Picardo,
Mitochondrial impairment in peripheral blood mononuclear cells during the active
phase of vitiligo, J. Invest. Dermatol. 117 (2001) 908–913.
[17] S.X. Huo, Q. Wang, X.M. Liu, C.H. Ge, L. Gao, X.M. Peng, M. Yan, The effect of butin
on the vitiligo mouse model induced by hydroquinone, Phytother. Res. 31 (2017)
740–746.
[18] H. Emily, W. Jonathan, D. Thuy, K. Reid, S. Lindsey, H. Quintin, H. Grace, C. Austin,
M. Lili, P. Stefan, Hydroxylated chalcones with dual properties: Xanthine oxidase
inhibitors and radical scavengers, Bioorgan. Med. Chem. 24 (2016) 578–587.
[19] M. Deodhar, D.S. Black, N. Kumar, Acid-catalyzed stereoselective rearrangement
and dimerization of flavenes: synthesis of dependensin, J. Cheminf. 63 (2007)
5227–5235.
[34] N.L. Vukovic, A.D. Obradovic, M.D. Vukic, D. Jovanovic, P.M. Djurjevic, Cytotoxic,
proapoptotic and antioxidative potential of flavonoids isolated from propolis
against colon (HCT-116) and breast (MDA-MB-231) cancer cell lines, Food. Res. Int.
106 (2018) 71–80.
[35] J. Gong, C. Cai, X. Liu, H. Jiang, D. Gao, H. Li, ChemMapper: a versatile web server
for exploring pharmacology and chemical structure association based on molecular
3D similarity method, Bioinformatics 29 (2013) 1827–1829.
[20] J.P. Lv, X.M. Zha, S.L. Pang, H.P. Jia, Y. Zhang, J. Shang, Synthesis and melano-
genesis evaluation of 3',4',7-trihydroxyflavanone derivatives and characterization of
flavanone-BODIPY, Bioorgan. Med. Chem. Lett. 25(2015), 1607–1610.
[21] J. Med. Chem. 55 (10) (2012) 4551–4567, https://doi.org/10.1021/jm201107g.
[22] T.Y. Choi, J.H. Kim, D.H. Ko, C.H. Kim, J.S. Hwang, S. Ahn, S.Y. Kim, C.D. Kim,
J.H. Lee, T.J. Yoon, Zebrafish as a new model for phenotype-based screening of
melanogenic regulatory compounds, Pigm. Cell. Res. 20 (2010) 120–127.
[23] E. Camp, M. Lardelli, Tyrosinase gene expression in zebrafish embryos, Dev. Genes.
Evol. 211 (2001) 150–153.
[36] F.F. Ping, J. Shang, J. Zhou, J. Song, L. Zhang, Activation of neurokinin-1 receptor
by substance P inhibits melanogenesis in B16–F10 melanoma cells, Int. J. Biochem.
Cell. B. 44 (2012) 2342–2348.
[37] J. Zhou, J.Y. Feng, W. Qian, J. Shang, Calcitonin gene-related peptide cooperates
with substance P to inhibit melanogenesis and induces apoptosis of B16F10 cells,
Cytokines 74 (2015) 137–144.
[38] S. Liao, J. Lv, J. Zhou, P.K. Kalavagunta, J. Shang, Effects of two chronic stresses on
mental state and hair follicle melanogenesis in mice, Exp. Dermatol. 26 (2017)
1083–1090.
[24] J. Oh, J. Kim, J.H. Jang, S. Lee, C.M. Park, W.K. Kim, J.S. Kin, Novel (1E,3E,5E)-1,6-
bis (Substituted phenyl) hexa-1,3,5-triene Analogs Inhibit Melanogenesis in B16F10
Cells and Zebrafish, Int. J. Mol. Sci. 19 (2018) 1067–1078.
533