Oxidation derivative of (-)-epigallocatechin-3-gallate (EGCG) inhibits RANKL-induced osteoclastogenesis by suppressing RANK signaling pathways in RAW 264.7 cells

Article abstract of DOI:10.1016/j.biopha.2019.109237

Tea consumption has positive effects on the skeletal system and prevents postmenopausal osteoporosis, mainly by inhibiting osteoclastogenesis. In green tea, (-)-epigallocatechin-3-gallate (EGCG) is the most abundant and active compound and has been shown to inhibit RANKL-induced osteoclast formation. Taking into account the highly oxidizable and unstable nature of EGCG, we hypothesized that EGCG oxidation product exhibits greater anti-osteoclastogenesis potential than EGCG. In this study, we successfully isolated and identified an EGCG oxidation derivative, (-)-gallocatechin gallate (compound 2), using a chemical oxidation strategy. We then compared the ability of compound 2 and EGCG to inhibit RANKL-induced osteoclastogenesis in RAW 264.7 cells. The results of TRAP staining and F-actin ring immunofluorescent staining showed that compound 2 exhibits stronger inhibition of RANKL-induced osteoclast differentiation and F-actin ring formation, respectively, than EGCG. Additionally, quantitative real-time PCR (qRT-PCR) and western blotting analyses showed that compound 2 significantly and more strongly inhibited the expression of osteoclastogenesis-related marker genes and proteins, including c-Src, TRAP, cathepsin K, β3-Integrin, and MMP-9, compared with EGCG. Furthermore, compound 2 significantly suppressed RANKL-induced expression of NFATc1 and c-Fos, the master transcriptional regulators of osteoclastogenesis, more strongly than EGCG. Mechanistically, molecular interaction assays showed that compound 2 binds to RANK with high affinity (K_D = 189 nM) and blocks RANKL–RANK interactions, thereby suppressing RANKL-induced early RANK signaling pathways including p65, JNK, ERK, and p38 in osteoclast precursors. Taken together, this study demonstrates for the first time that an oxidation derivative of EGCG (compound 2) inhibits RANKL-induced osteoclastogenesis by suppressing RANK signaling pathways in RAW 264.7 cells.

Full text of DOI:10.1016/j.biopha.2019.109237

Biomedicine&Pharmacotherapy118(2019)109237
Contents lists available at ScienceDirect
Biomedicine & Pharmacotherapy
journal homepage: www.elsevier.com/locate/biopha
Oxidation derivative of (-)-epigallocatechin-3-gallate (EGCG) inhibits
RANKL-induced osteoclastogenesis by suppressing RANK signaling
pathways in RAW 264.7 cells
Huanhuan Xua,b,1, Titi Liua,c,1, Jin Lia,c,1, Jing Xua,c, Fei Chena,c, Lihong Hua,c, Banglei Zhanga,c
,
⁎
⁎
⁎
Chengting Zia,b, , Xuanjun Wanga,b,d, , Jun Shenga,d,
a Key Laboratory of Pu-er Tea Science, Ministry of Education, Yunnan Agricultural University, Kunming, 650201, China
b College of Science, Yunnan Agricultural University, Kunming, 650201, China
c College of Food Science and Technology, Yunnan Agricultural University, Kunming, 650201, China
d State Key Laboratory for Conservation and Utilization of Bio-Resources in Yunnan, Kunming, 650201, China
A R T I C L E I N F O
A B S T R A C T
Keywords:
EGCG
Oxidation derivative
Osteoclastogenesis
NFATc1
Tea consumption has positive effects on the skeletal system and prevents postmenopausal osteoporosis, mainly
by inhibiting osteoclastogenesis. In green tea, (-)-epigallocatechin-3-gallate (EGCG) is the most abundant and
active compound and has been shown to inhibit RANKL-induced osteoclast formation. Taking into account the
highly oxidizable and unstable nature of EGCG, we hypothesized that EGCG oxidation product exhibits greater
anti-osteoclastogenesis potential than EGCG. In this study, we successfully isolated and identified an EGCG
oxidation derivative, (-)-gallocatechin gallate (compound 2), using a chemical oxidation strategy. We then
compared the ability of compound 2 and EGCG to inhibit RANKL-induced osteoclastogenesis in RAW 264.7 cells.
The results of TRAP staining and F-actin ring immunofluorescent staining showed that compound 2 exhibits
stronger inhibition of RANKL-induced osteoclast differentiation and F-actin ring formation, respectively, than
EGCG. Additionally, quantitative real-time PCR (qRT-PCR) and western blotting analyses showed that compound
2 significantly and more strongly inhibited the expression of osteoclastogenesis-related marker genes and pro-
teins, including c-Src, TRAP, cathepsin K, β3-Integrin, and MMP-9, compared with EGCG. Furthermore, com-
pound 2 significantly suppressed RANKL-induced expression of NFATc1 and c-Fos, the master transcriptional
regulators of osteoclastogenesis, more strongly than EGCG. Mechanistically, molecular interaction assays
showed that compound 2 binds to RANK with high affinity (KD = 189 nM) and blocks RANKL–RANK interac-
tions, thereby suppressing RANKL-induced early RANK signaling pathways including p65, JNK, ERK, and p38 in
osteoclast precursors. Taken together, this study demonstrates for the first time that an oxidation derivative of
EGCG (compound 2) inhibits RANKL-induced osteoclastogenesis by suppressing RANK signaling pathways in
RAW 264.7 cells.
c-Fos
RANK signaling pathways
1. Introduction
periodontal disease, multiple myeloma, and metastatic cancers [1,3].
Thus, inhibiting osteoclast differentiation and activation is a key ther-
Bone homeostasis is tightly regulated by the coupled action of os-
teoclasts that regulate bone resorption and osteoblasts that regulate
bone formation via a process called remodeling [1]. An imbalance be-
tween the resorption of old bones and formation of new bones directly
causes gross perturbations in the skeletal structure and function, po-
tentially leading to morbidity and shortened lifespan [1,2]. Excess os-
teoclastic activity contributes to the development of many skeletal
diseases, such as postmenopausal osteoporosis, rheumatoid arthritis,
aeputic strategy for treating bone metabolic diseases [4].
Osteoclasts originate from hematopoietic stem cells and differ-
entiate through monocyte and macrophage precursors to form mature
osteoclasts, which are characterized by a clear zone, ruffled border, and
the secretion of acid and lytic enzymes that degrade bone mineralized
matrices [5,6]. The receptor activator of NF-κB (RANK) ligand (RANKL)
plays an important role in osteoclastogenesis [7]. The binding of
RANKL to its receptor RANK directly activates canonical RANK
⁎ Corresponding authors at: Yunnan Agricultural University, Heilongtan, North of Kunming, Kunming, 650201, China.
E-mail addresses: zichengting@126.com (C. Zi), wangxuanjun@gmail.com (X. Wang), shengjunpuer@163.com (J. Sheng).
1 These authors have contributed equally to this work.
https://doi.org/10.1016/j.biopha.2019.109237
Received 22 May 2019; Received in revised form 13 July 2019; Accepted 16 July 2019
0753-3322/©2019TheAuthors.PublishedbyElsevierMassonSAS.ThisisanopenaccessarticleundertheCCBY-NC-NDlicense
(http://creativecommons.org/licenses/BY-NC-ND/4.0/).

H. Xu, et al.
Biomedicine&Pharmacotherapy118(2019)109237
signaling pathways during osteoclastogenesis, leading to the activation
of many key transcription factors such as c-Fos and NFATc1 (short for
Nuclear Factor of Activated T-Cells, Cytoplasmic1). These transcription
factors regulate the expression of osteoclast-related marker genes such
as tartrate-resistant acid phosphatase (TRAP), c-Src, cathepsin K, matrix
metallopeptidase-9 (MMP-9), and c-Src, thus driving the formation and
activation of mature osteoclasts [6,8,9].
Louis, MO, USA). Specific primary antibodies against NFATc1, c-Src, c-
Fos, cathepsin K, and JNK were purchased from Santa Cruz
Biotechnology (Santa Cruz, CA, USA). Antibodies against phosphory-
lated p65 (p-p65), p65, p-JNK, p-ERK1/2, ERK1/2, p-p38, and p38
were purchased from Cell Signaling Technology (Beverly, MA, USA).
Anti-TRAP, anti-β-tubulin, and horseradish peroxidase-conjugated sec-
ondary antibodies were purchased from Abcam (Cambridge, MA, USA),
Proteintech Group, Inc. (IL, USA), and R&D Systems (Minneapolis, MN,
USA), respectively.
Tea prepared from the green leaves of Camellia sinensis (L.) O.
Kuntze is the second most consumed beverage worldwide, given its
broad-spectrum beneficial effects on human health [10]. Tea is cate-
gorized into three types: green, black, and Pu-erh tea, depending on the
processing technique [11]. Numerous studies have shown that tea
consumption has positive effects on the skeletal system by effectively
preventing and treating postmenopausal osteoporosis [12]. Catechins
including (-)-epigallocatechin-3-gallate (EGCG), (-)-epicatechin-3-gal-
late (ECG), (-)-epigallocatechin (EGC), and (-)-epicatechin (EC) are the
main active components of green tea and have diverse beneficial effects
on human health [13]. Among these, EGCG has recently attracted great
interest for its antiviral, anti-inflammatory, and anticancer properties
[13–16]. Furthermore, EGCG has been shown to inhibit RANKL-in-
duced osteoclast formation [17,18]. However, since EGCG is easily
oxidized and is unstable at high pH or under prolonged exposure to
high temperature [19,20], we hypothesized that oxidation products of
EGCG would exhibit higher anti-osteoclastogenesis potential than
EGCG.
2.2. Reaction components and extraction
EGCG (0.45 mmol) was maintained in 10 mL of 0.05 M phosphate
buffer (Na2HPO4, pH 5.8) at 60 °C for 2 h. The reaction mixture was
extracted with ethyl acetate (3 × 10 mL), and the combined organic
layer was dried over sodium sulfate. The solvent was evaporated under
vacuum to obtain the residue.
2.3. HPLC analysis and isolation
Analysis of EGCG and the reaction mixture was performed on an
Agilent 1260 LC system, with an XTerra RP-18 (4.6 mm × 250 mm)
column. Absorption was detected at 254 nm. The injecting volume was
10 μL, and the flow rate was 1 mL/min. The ratio of mobile phase A
(methanol) to phase
B (0.5% acetic acid in water) was 20:80.
In this study, we isolated and identified an EGCG oxidation deri-
vative, (-)-gallocatechin gallate (GCG; compound 2), using a chemical
oxidation strategy. We tested the ability of compound 2 with EGCG to
inhibit RANKL-induced osteoclast differentiation in RAW 264.7 cells.
We also explored the potential molecular mechanisms underlying the
inhibition of RANKL-induced osteoclast differentiation.
Separation and purification of the compounds were performed on an
Agilent 1260 LC system, with an XTerra RP-18 (9.8 mm × 250 mm)
column, using 20% methanol solution containing 0.5% acetic acid.
2.4. Characteristics of compound 2
Compound 2 was a white amorphous powder, m.p. 220–222 °C, 1H-
NMR (CD3OD, 500 MHz) and 13C-NMR (CD3OD, 125 MHz) data
(Table 1); negative-ion ESIMS; m/z 457 [M−H]−; negative-ion HRE-
SIMS was calculated for C22H17O11 [M−H]− 457.0849 and was found
to be 457.0850.
2. Materials and methods
2.1. Materials
High purity (> 98%) grade EGCG was purchased from Biopurify
Phytochemicals Ltd. (Chengdu, China). High performance liquid chro-
matography (HPLC) grade methanol was obtained from J&K Chemical
Technology Co., Ltd. (Beijing, China). All reagents were commercially
available and used without further purification, unless indicated
otherwise. All solvents were obtained from commercial sources and
were purified according to standard procedures.
Melting points (compound 2) were measured using an X-4 melting
point apparatus. Mass spectra (MS) and high-resolution mass spectra
(HRMS) were recorded using Agilent 6540 Q-Tof (ESIMS) (Agilent, CA,
USA). Proton nuclear magnetic resonance (1H-NMR) and carbon-13
NMR (13C-NMR) spectra were recorded using Bruker AVANCE III
500 MHz instruments (Bruker BioSpin GmbH, Rheinstetten, Germany),
with tetramethylsilane (TMS) as an internal standard. To perform
column chromatography and thin-layer chromatography (TLC), silica
gel (200–300 mesh) and silica GF254, respectively, were produced by
Qingdao Marine Chemical Company (China).
Recombinant mouse RANKL (R&D Systems, Minneapolis, MN, USA)
was reconstituted to a final concentration of 10 μg/mL using sterile
phosphate-buffered saline (PBS) containing 0.1% bovine serum al-
bumin (BSA) and then stored at −20 °C. The prepared stock solution
was added to the culture medium to achieve the indicated final con-
centration. Recombinant mouse RANK protein for molecular interaction
assay was purchased from R&D Systems. Dulbecco’s Modified Eagle’s
Medium (DMEM) and fetal bovine serum (FBS) were sourced from
Hyclone Laboratories (Logan, UT, USA) and Biological Industries
(Kibbutz Beit Haemek, Israel), respectively. Mixed penicillin–strepto-
mycin solution (P/S), TRITC-conjugated phalloidin, and antifade
mounting medium with DAPI were purchased from Solarbio (Beijing,
China). The TRAP staining kit was purchased from Sigma-Aldrich (St.
2.5. Cell culture
Mouse macrophage RAW 264.7 cells obtained from American Type
Table 1
1H-NMR and 13C-NMR data for (-)-gallocatechin gallate (GCG) identified pre-
viously and compound 2 identified in this study.
GCG (δ1)a
Compound 2 (δ2)
C position δC δH
δC
δH
2
3
77.2
69.2
5.52 (d, J =4.8 Hz)
4.96 (dd, J =4.8 Hz,
10.8 Hz)
2.59 (s, H-4α)
2.51 (dd, J =2.0 Hz,
3.6 Hz, H-4β)
78.6
69.9
5.52 (d, J =2.4 Hz)
4.96 (d, J =2.0 Hz)
4
22.6
26.8
2.59 (s, H-4α)
2.51 (dd, J =2.0 Hz,
3.6 Hz, H-4β)
5 (7)
6
156.5
95.7
157.9
96.4
95.8
5.93 (d, J =2.4 Hz)
5.83 (d, J =2.0 Hz)
5.93 (s)
5.93 (s)
8
94.3
9
10
154.9
97.5
157.9
99.4
130.7
106.8 6.48 (s)
146.7
133.8
121.4
110.2 6.93 (s)
146.7
139.7
167.6
1′
129.0
2′, 6′
3′, 5′
4′
105.2 6.27 (s)
146.2
132.9
119.5
108.9 6.86 (H, s)
145.8
1''
2'', 6''
3'', 5''
4''
139.0
C = O
165.5
a
Experimental data for GCG were obtained from a previous study [27].
2

H. Xu, et al.
Biomedicine&Pharmacotherapy118(2019)109237
Culture Collection (ATCC; TIB-71™; Manassas, VA, USA) were cultured
in DMEM supplemented with 10% FBS and 1% P/S. The cells were
incubated at 37 °C in a humidified atmosphere of 95% air and 5% CO2.
60 °C for 30 s. Primers used for qRT-PCR were selected from previous
studies [21,22] and are listed in Table 2. All reactions were performed
in triplicate in a volume of 10 μL. Relative mRNA expression was de-
termined using the comparative 2−ΔΔCT method and normalized re-
lative to the GAPDH gene (endogenous control).
2.6. In vitro osteoclastogenesis assay
RANKL induces the differentiation of RAW 264.7 cells into mature
osteoclasts in vitro, as previously described [21]. To perform the os-
teoclastogenesis assay, RAW 264.7 cells were seeded in 96-well plates
at a density of 1 × 103 cells per well and allowed to adhere overnight.
The osteoclast precursor cells were pretreated with or without EGCG
(compound 1, 10 μM) or compound 2 (10 μM) for 20 min in the specific
osteoclastogenic medium (DMEM supplemented with 6% FBS and 1%
P/S) and subsequently stimulated with RANKL (50 ng/mL) for 4 days.
TRAP-positive multinuclear cells were visualized using a TRAP staining
kit, according to the manufacturer’s instructions. TRAP-positive multi-
nuclear cells with more than three nuclei in each well were counted as
mature osteoclasts under a light microscope (Olympus, Tokyo, Japan),
and images were acquired with a microscope-aided camera.
2.9. Western blot analysis
Cells were washed twice with ice-cold PBS and lysed using RIPA
buffer (Solarbio) on ice for protein extraction. The protein lysates were
quantified using an Enhanced BCA Protein Assay Kit (Beyotime
Biotechnology, Shanghai, China), according to the manufacturer’s in-
structions. To separate the isolated proteins, sodium dodecyl sulfate-
polyacrylamide gel electrophoresis (SDS-PAGE) was performed using
equal amounts of proteins, which were then transferred to poly-
vinylidene fluoride (PVDF) membranes (Millipore, Billerica, USA).
Proteins of interest were detected by immunoblotting using specific
primary antibodies, followed by the corresponding horseradish perox-
idase-conjugated secondary antibodies. Hybridization of antibodies to
the proteins was detected using an Ultra-sensitive Enhanced
Chemiluminescent Substrate Kit (Beijing 4A Biotech Co., Ltd; Beijing,
China), and images were captured using a FluorChem E System
(ProteinSimple, San Jose, CA, USA). Band densities were quantified
using AlphaView software (Cell Biosciences, Santa Clara, CA, USA).
2.7. F-actin ring immunofluorescent staining
RAW 264.7 cells (2 × 104 cells per well) were cultured on glass
coverslips in a 12-well plate and incubated overnight. The cells were
pretreated with or without compound 1 (10 μM) or compound 2
(10 μM) for 20 min in the specific osteoclastogenic medium and sub-
sequently stimulated with RANKL (50 ng/mL) for 6 days. The cells were
washed, fixed in 4% paraformaldehyde for 10 min, and permeabilized
with 0.1% Triton X-100 for 5 min. Subsequently, the cells were in-
cubated in 150 nM TRITC-conjugated phalloidin at 37 °C for 30 min and
then washed with PBS. Finally, the cells were mounted using antifade
mounting medium with DAPI for 2 min. F-actin ring distribution was
visualized using DM2000 fluorescence microscope (Leica). Images were
captured at a magnification of 200X and merged using the ImageJ
software (National Institutes of Health, Bethesda, USA).
2.10. Molecular interaction assay
Interaction between mouse RANK protein and compound 2 was
investigated using biolayer interferometry (BLI) analysis. RANK was
biotinylated using amine-PEG3-biotin, as described previously [23].
Then, the reaction mixture was desalted using Zeba Spin desalting
columns, and the biotinylated RANK protein was immobilized onto the
surface of Super Streptavidin (SSA) biosensors. Increasing concentra-
tions of compound 2 were allowed to interact with the immobilized
RANK at 30 °C in PBS (pH 7.4) using an Octet Red 96 instrument
(Fortebio, USA). The final volume of all solutions was 200 μL. Assays
were performed in black solid 96-well flat bottom plates on a shaker set
at 1000 r/min. The association and dissociation of compound 2 with
RANK was measured for 600 s each. Kinetic parameters and affinities
were calculated from a non-linear global fit of the data between com-
pound 2 and RANK using Octet Data Analysis software version 7.0
(Fortebio).
To test whether compound 2 competes with RANKL to bind to its
receptor RANK, a solution competition BLI study was performed using
the Octet Red 96 instrument. Briefly, compound 2 (0 or 10 μM) was
preincubated with immobilized RANK for 720 s. Then, 0.4 μM RANKL
containing 0 or 10 μM compound 2 was allowed to interact with im-
mobilized RANK for 420 s. Finally, dissociation was followed for
420–840 s using 0 or 10 μM compound 2.
2.8. RNA extraction and quantitative real-time PCR (qRT-PCR) analysis
RAW 264.7 cells (1 × 105 cells per well) were seeded in 12-well
plates and allowed to adhere overnight. Subsequently, RAW 264.7 cells
were pretreated with or without compound 1 (10 μM) or compound 2
(10 μM) for 20 min in the specific osteoclastogenic medium and then
stimulated with RANKL (50 ng/mL) for 60 h. Then, total RNA was ex-
tracted from RAW 264.7 cells using TransZol™ Up Reagent (TransGen
Biotech, Beijing, China), according to the manufacturer’s protocol.
First-strand cDNA was synthesized from the isolated RNA (1 μg) using
PrimeScript™ RT Reagent Kit with gDNA Eraser (TaKaRa Bio, Otsu,
Japan), according to the manufacturer’s instructions. Subsequently,
qRT-PCR was performed on a 7900 H T Fast Real-Time PCR system
(Applied Biosystems, Foster City, CA, USA) using the TB Green® Premix
Ex Taq™ II (Tli RNaseH Plus) Reagent (TaKaRa) under the following
conditions: 1 cycle of initial denaturation at 95 °C for 30 s, followed by
40 cycles of denaturation at 95 °C for 5 s and annealing and extension at
2.11. Statistical analysis
All experimental data were expressed as mean
standard error of
Table 2
List of primers used for quantitative real-time PCR (qRT-PCR).
Gene
Forward primer (5′→3′)
Reverse primer (5′→3′)
GAPDH
c-Src
TRAP
cathepsin K
β3-Integrin
MMP-9
NFATc1
c-Fos
AACTTTGGCATTGTGGAAGG
CCAGGCTGAGGAGTGGTACT
GCTGGAAACCATGATCACCT
CTTCCAATACGTGCAGCAGA
TGACATCGAGCAGGTGAAAG
CGTCGTGATCCCCACTTACT
TGGAGAAGCAGAGCACAGAC
CAAGCGGAGACAGATCAACTTG
ACACATTGGGGGTAGGAACA
CAGCTTGCGGATCTTGTAGT
GAGTTGCCACACAGCATCAC
TCTTCAGGGCTTTCTCGTTC
GAGTAGCAAGGCCAATGAGC
AACACACAGGGTTTGCCTTC
GCGGAAAGGTGGTATCTCAA
TTTCCTTCTCTTTCAGCAGATTGG
3

H. Xu, et al.
Biomedicine&Pharmacotherapy118(2019)109237
Fig. 1. High performance liquid chromatography (HPLC) analysis of the reaction of (-)-epigallocatechin-3-gallate (EGCG) with phosphate buffer. (A) EGCG. (B)
Reaction of EGCG with 0.5 M phosphate buffer (Na2HPO4; pH 5.8) at 60 °C for 2 h.
the mean (SEM) of at least three independent experiments. Statistical
analysis was performed using the Student’s t-test and differences at
P < 0.05 were considered statistically significant. All analyses were
performed using SPSS 17.0 (Chicago, IL, USA) and GraphPad Prism 5
(GraphPad Software, Inc., La Jolla, CA, USA).
3.2. Compound 2 is a stronger inhibitor of osteoclast differentiation than
EGCG
Numerous studies suggest that tea and its active components,
especially catechins, prevent and treat osteoporosis by improving bone
mineral density and stimulating osteoblast differentiation, while sup-
pressing osteoclast formation and differentiation [28–30]. EGCG and its
derivatives have been reported to exhibit positive effects on bone me-
tabolism [31,32]. However, comparative studies on the anti-osteoclas-
togenesis activity of EGCG and its oxidation derivative are relatively
limited. To investigate whether the inhibitory effect of compound 2
(GCG) on RANKL-induced osteoclast differentiation was comparable
with that of EGCG, we treated RAW 264.7 cells with RANKL (50 ng/mL)
in the presence of EGCG or compound 2 (10 μM) for 4 days, and TRAP
staining was performed to detect osteoclast formation. RAW 264.7 cells
differentiated into mature osteoclasts after RANKL stimulation
(P < 0.001; Fig. 3). When 10 μM of compound 2 was added to the re-
action, the RANKL-induced osteoclast formation was almost completely
inhibited. However, when EGCG was added to the reaction at the same
concentration (10 μM), osteoclast differentiation was only partially in-
hibited (Fig. 3), which is consistent with the result of a previous study
[33]. Compared with EGCG treatment, the number of TRAP-positive
multinuclear osteoclast cells decreased significantly following treat-
ment with compound 2 (P < 0.01; Fig. 3). These results suggest that
compound 2 is a stronger inhibitor of RANKL-induced osteoclast dif-
ferentiation than EGCG.
3. Results and discussion
3.1. Chemical oxidation and structural determination of EGCG
Chemical oxidation of EGCG has been reported using various oxi-
dative compounds, including K3[Fe(CN)6] [24], CuSO4 [25], and CuCl2
[25], that triggered the formation of dimers such as theaflavins, thea-
sinensins, and oolongtheanins and other polymers. Generally, EGCG is
oxidized in solution (indicated by changing colors), leading to the
formation of an isomer, (-)-gallocatechin gallate (GCG), under acidic
catalytic conditions [26,27]. In this study, the chemical oxidation of
EGCG was based on the method reported by Xie et al. [27]. EGCG was
dissolved in 0.05 M Na2HPO4 (pH 5.8) at 60 °C for 2 h, and the mixture
was analyzed by HPLC. Interestingly, compound 2 was obtained as the
major product on the HPLC spectrum (Fig. 1), with a yield of 60% after
extraction and evaporation.
Compound 2 was obtained as a white amorphous powder. One
[MeH]− peak on the MS spectrum demonstrated that the molecular
weight of compound 2 was 457, implying that it was an EGCG ana-
logue. On the other hand, in the 1H-NMR spectra (Table 1), the proton
at C-2 (5.52 ppm, d, J =2.4 Hz) and C-3 (4.96 ppm, d, J =2.0 Hz) ap-
peared as a double peak, indicating a cis-relationship between H-2 and
H-3; this product was identified as GCG [27]. The bond between C-2
and O atom was broken under acid catalysis, and the carbon atom at the
C-2 position formed tertiary carbocation transition states, generating
GCG (Fig. 2).
3.3. Compound 2 is a stronger inhibitor of RANKL-stimulated F-actin ring
formation than EGCG
Mature osteoclasts secrete several proteinases and pump out protons
to resorb the bone and form a characteristic actin ring structure to
confine the acidic resorbing microenvironment [34]. In this study,
mature osteoclasts formed many characteristic circular F-actin rings in
the periphery of cells after RANKL stimulation (P < 0.001; Fig. 4).
Fig. 2. Schematic representation of a plausible mechanism for the conversion of EGCG to GCG.
4

H. Xu, et al.
Biomedicine&Pharmacotherapy118(2019)109237
Fig. 3. Compound 2 inhibits osteoclast differ-
entiation more effectively than EGCG. RAW
264.7 cells were pretreated with or without
compound 1 (EGCG, 10 μM) or compound 2
(10 μM) for 20 min in an osteoclastogenic
medium and subsequently stimulated with
RANKL (50 ng/mL) for
4 days. Cells were
stained with TRAP; TRAP-positive multi-
nuclear cells with more than three nuclei in
each well were counted under a microscope
(original magnification 40X). Representative
images are shown. Data represent mean
standard error of mean (SEM) of three in-
dependent replicates. ***P < 0.001 compared
with the control; ###P < 0.001 compared
with RANKL treatment only; $$P < 0.01
compared with compound 1 treatment.
However, treatment with EGCG or compound 2 significantly suppressed
the formation of F-actin rings (P < 0.001). Unsurprisingly, compound 2
inhibited RANKL-stimulated F-actin ring formation more strongly than
EGCG (P < 0.01).
effectively than EGCG (P < 0.001; Fig. 5F–H), which was consistent
with the results of qRT-PCR analysis.
3.5. Compound 2 significantly suppresses RANKL-induced expression of
NFATc1 and c-Fos more than EGCG
3.4. Compound 2 significantly inhibits the expression of osteoclastogenesis-
related marker genes and proteins more than EGCG
Many transcription factors are essential for promoting osteoclast-
related gene expression during osteoclastogenesis [8,9]. To understand
the inhibitory mode of action of compound 2 on RANKL-induced os-
teoclast differentiation, the expression of NFATc1 and c-Fos, master
transcriptional regulators of osteoclastogenesis [34], were examined
using qRT-PCR and western blotting analysis. The results showed that
mRNA and protein expression levels of NFATc1 and c-Fos were sig-
nificantly increased by the stimulation of RANKL (P < 0.001; Fig. 6)
but significantly decreased following treatment with EGCG or com-
pound 2 (P < 0.05). In addition, compound 2 significantly suppressed
the expression of these key transcription factors more than EGCG
(P < 0.05; Fig. 6). On the basis of these results, we confirmed the in-
hibitory effect of the EGCG oxidation derivative (compound 2) on
RANKL-induced osteoclastogenesis in RAW 264.7 cells.
Osteoclast differentiation and activation are executed directly by
the expression of a large number of marker genes, such as c-Src, TRAP,
cathepsin K, β3-Integrin, and MMP-9 [8,21]. To further confirm that
compound 2 inhibits RANKL-induced osteoclastogenesis more effec-
tively than EGCG, we examined the expression of osteoclastogenesis-
specific marker genes and proteins after treatment with EGCG or
compound 2. The results of qRT-PCR analysis demonstrated that the
mRNA expression levels of c-Src, TRAP, cathepsin K, β3-Integrin, and
MMP-9 were significantly increased by the stimulation of RANKL
(P < 0.001; Fig. 5A–E) but significantly down-regulated after treatment
with EGCG or compound 2 (P < 0.01). Compared with EGCG treat-
ment, the expression of these osteoclastogenesis-specific marker genes
significantly decreased following compound 2 treatment (P < 0.001).
Additionally, EGCG or compound 2 significantly suppressed the ex-
pression of osteoclast-related marker proteins, including c-Src, TRAP,
and cathepsin K (P < 0.01; Fig. 5F–I). Furthermore, compound 2 in-
hibited the protein expression of c-Src and TRAP significantly more
3.6. Compound 2 inhibits RANKL-induced phosphorylation of p65 and
MAPK in RAW 264.7 cells
To further illuminate the molecular mechanisms underlying the
Fig. 4. Compound 2 inhibits RANKL-stimulated F-actin ring formation more strongly than EGCG. RAW 264.7 cells were pretreated with or without compound 1
(EGCG, 10 μM) or compound 2 (10 μM) for 20 min in an osteoclastogenic medium and subsequently stimulated with RANKL (50 ng/mL) for 6 days. F-actin rings and
nuclei were stained with TRITC-conjugated phalloidin and DAPI, respectively. Representative images are displayed (original magnification 200X); images were
merged using the ImageJ software. Data represent mean
SEM of three independent replicates. ***P < 0.001 compared with the control; ###P < 0.001 compared
with RANKL treatment only; $$P < 0.01 compared with compound 1 treatment.
5

H. Xu, et al.
Biomedicine&Pharmacotherapy118(2019)109237
Fig. 5. Compound 2 significantly inhibits the expression of osteoclastogenesis-related marker genes and proteins more than EGCG. RAW 264.7 cells were pretreated
with or without compound 1 (EGCG, 10 μM) or compound 2 (10 μM) for 20 min in an osteoclastogenic medium and subsequently stimulated with RANKL (50 ng/mL)
for 60 h. (A–E) Analysis of c-Src (A), TRAP (B), cathepsin K (C), β3-Integrin (D), and MMP-9 (E) genes by qRT-PCR. (F) Western blotting analysis of c-Src, TRAP, and
cathepsin K proteins. Representative images are shown. (G–I) Gray densities of bands corresponding to c-Src (G), TRAP (H), and cathepsin K (I) quantified using
AlphaView software. Data represent mean
SEM of three independent replicates. ***P < 0.001 compared with the control; ##P < 0.01 and ###P < 0.001
compared with RANKL treatment only; $$$P < 0.001 compared with compound 1 treatment. ns, not significant.
inhibitory effects of compound 2 on RANKL-induced osteoclastogenesis
and NFATc1 and c-Fos expression, we examined the effects of com-
pound 2 on RANKL-induced early canonical RANK signaling pathways,
including NF-κB and MAPK (JNK, ERK, and p38), which activate key
transcription factors such as NFATc1 and c-Fos during osteoclastogen-
esis [8,9]. Phosphorylation of these signaling molecules was observed
in RAW 264.7 cells at 5 min after RANKL treatment (Fig. 7A). Within
60 min of stimulation with RANKL, phosphorylation of p65 (P < 0.01;
Fig. 7B), JNK (P < 0.01; Fig. 7C), ERK1/2 (P < 0.01; Fig. 7D), and p38
(P < 0.05; Fig. 7E) relative to total p65, total JNK, total ERK1/2, and
total p38 were significantly suppressed by compound 2 treatment in
RAW 264.7 cells. Collectively, these results indicate that compound 2
exerts an inhibitory effect on RANKL-stimulated NF-κB and MAPK ac-
tivation pathways.
3.7. Compound 2 binds to RANK with high affinity and blocks
RANKL–RANK interactions
Protein–protein interactions between RANKL and RANK are known
to manipulate the differentiation and activation of osteoclasts during
bone remodeling and are therefore considered a pivotal therapeutic
target for the treatment of osteoporosis [35]. Given the inhibitory ef-
fects of compound 2 on RANKL-induced early RANK signaling path-
ways including p65, JNK, ERK, and p38, we further examined whether
compound 2 blocks RANKL–RANK interactions. First, the binding affi-
nity of compound 2 for RANK was detected using BLI analysis (Fig. 8A).
The results showed that compound 2 directly interacts with RANK. The
apparent association (Kon) and dissociation (Koff) constants of com-
pound 2 were estimated as 2.27 × 102
M ,
−1s−1 and 4.28 × 10-5 s−1
6

H. Xu, et al.
Biomedicine&Pharmacotherapy118(2019)109237
Fig. 6. Compound 2 significantly suppresses
RANKL-induced expression of NFATc1 and c-
Fos more than EGCG. Treatment conditions
were the same as described in Fig. 5. (A, B)
Expression analysis of NFATc1 (A) and c-Fos
(B) by qRT-PCR. (C) Western blotting analysis
of NFATc1 and c-Fos proteins. Representative
images are shown. (D, E) Gray densities of
bands corresponding to NFATc1 (D) and c-Fos
(E) quantified using AlphaView software. Data
represent mean
SEM of three independent
replicates. ***P < 0.001 compared with the
control; #P < 0.05, ##P < 0.01, and
###P < 0.001 compared with RANKL treat-
ment only; $P < 0.05 and $$$P < 0.001
compared with compound 1 treatment.
respectively, and RANK-binding affinity (KD) was estimated as 189 nM
(Fig. 8A). These data suggest that the RANK–compound 2 complex is
relatively stable. Next, solution competition BLI assay revealed that
compound 2 has the ability to compete with RANKL to bind to its re-
ceptor RANK (Fig. 8B), suggesting that compound 2 blocks RANKL–-
RANK interactions. These results suggest the receptor RANK as a po-
tential direct interaction target of compound 2 in the inhibition of
osteoclastogenesis, although this needs further verification in the fu-
ture.
To maintain bone homeostasis, a balance must be achieved between
bone resorption and bone formation [1]. Tea catechin EGCG has been
shown to enhance osteoblast differentiation [36]. However, whether
the oxidation derivative of EGCG, compound 2, could stimulate
osteoblast differentiation needs further investigation. Additionally,
there is an urgent need to investigate the potential anti-osteoporosis
effects of compound 2 in vivo.
4. Conclusion
In this study, we demonstrated for the first time that an EGCG
oxidation derivative (compound 2) inhibits RANKL-induced osteoclas-
togenesis and F-actin ring formation more effectively than EGCG
through the down-regulation of NFATc1 and c-Fos. Compound 2
showed binding to RANK with high affinity (KD = 189 nM) and blocked
RANKL–RANK interactions, thereby suppressing RANKL-induced early
RANK signaling pathways including p65, JNK, ERK, and p38 in
7

H. Xu, et al.
Biomedicine&Pharmacotherapy118(2019)109237
Fig. 7. Compound 2 suppresses RANKL-stimulated phosphorylation of p65 and MAPK. (A) RAW 264.7 cells were pretreated with or without compound 2 (10 μM) for
3 h in serum-free medium and then stimulated with RANKL (50 ng/mL) for the indicated times. Cell lysates were subjected to western blotting using the indicated
primary antibodies. Representative images are shown. (B–E) Quantification of the ratios of band intensity relative to total p65 (B), total JNK (C), total ERK1/2 (D),
and total p38 (E). Data are expressed as mean
SEM of three independent experiments. Asterisks indicate significant differences compared with the RANKL
treatment at the same time point (*P < 0.05; **P < 0.01; ***P < 0.001).
Fig. 8. Compound 2 binds to RANK with high affinity and blocks RANKL–RANK interactions. (A) BLI sensorgrams indicating the interactions between gradient
concentrations of compound 2 and RANK were measured on an Octet Red 96 system, with association and dissociation for 600 s each. These sensorgrams were used
to determine the binding affinity (KD), association constant (Kon), and dissociation constant (Koff). (B) Compound 2 competes with RANKL to bind to its receptor
RANK in the solution competition BLI assay. Compound 2 (0 or 10 μM) was preincubated with 0.4 μM RANKL prior to interaction with immobilized RANK.
Sensorgrams correspond to the phases of association and dissociation for 420 s each.
8

H. Xu, et al.
Biomedicine&Pharmacotherapy118(2019)109237
osteoclast precursors.
(1) (2018).
[14] J. Xu, Z. Xu, W. Zheng, A review of the antiviral role of green tea catechins,
Molecules 22 (8) (2017).
Author contributions
[15] A. Negri, V. Naponelli, F. Rizzi, S. Bettuzzi, Molecular targets of epigallocatechin-
gallate (EGCG): a special focus on signal transduction and cancer, Nutrients 10 (12)
(2018).
[16] T. Wang, Z. Xiang, Y. Wang, X. Li, C. Fang, S. Song, C. Li, H. Yu, H. Wang, L. Yan,
S. Hao, X. Wang, J. Sheng, (-)-Epigallocatechin gallate targets notch to attenuate the
inflammatory response in the immediate early stage in human macrophages, Front.
Immunol. 8 (2017) 433.
[17] S.T. Chen, L. Kang, C.Z. Wang, P.J. Huang, H.T. Huang, S.Y. Lin, S.H. Chou, C.C. Lu,
P.C. Shen, Y.S. Lin, C.H. Chen, (-)-Epigallocatechin-3-Gallate Decreases
Osteoclastogenesis via Modulation of RANKL and Osteoprotegrin, Molecules 24 (1)
(2019).
[18] S.H. Lee, B.J. Kim, H.J. Choi, S.W. Cho, C.S. Shin, S.Y. Park, Y.S. Lee, S.Y. Lee,
H.H. Kim, G.S. Kim, J.M. Koh, (-)-Epigallocathechin-3-gallate, an AMPK activator,
decreases ovariectomy-induced bone loss by suppression of bone resorption, Calcif.
Tissue Int. 90 (5) (2012) 404–410.
Jun Sheng, Xuanjun Wang, Huanhuan Xu, and Chengting Zi con-
ceived and designed the experiments. Huanhuan Xu, Titi Liu, Jin Li,
Jing Xu, Chengting Zi, Fei Chen, Lihong Hu, and Banglei Zhang per-
formed the experiments. Huanhuan Xu and Chengting Zi analyzed the
data. Jun Sheng and Xuanjun Wang contributed reagents/materials/
analysis tools. Huanhuan Xu, Titi Liu, and Chengting Zi wrote the
manuscript. Huanhuan Xu, Titi Liu, and Jin Li revised the manuscript.
All authors read and approved the final version manuscript.
Funding
[19] M. Friedman, H.S. Jurgens, Effect of pH on the stability of plant phenolic com-
pounds, J. Agric. Food Chem. 48 (6) (2000) 2101–2110.
This study was supported in part by research grants from the
Scholarship for Academic Leader of Yunnan Province (2017HA015), the
Major Scientific and Technological Special Project of Yunnan Province
(2017ZF003, 2018ZG010 and 2018ZG013), the National Natural
Science Foundation of China (21602196), and the National Natural
Science Foundation of Yunnan Province (2017FD084). The funders had
no role in the study design, data collection and analysis, decision to
publish, or preparation of the manuscript.
[20] J. Wang, H. Tang, B. Hou, P. Zhang, J. Sheng, Synthesis, antioxidant activity, and
density functional theory study of catechin derivatives, RSC Adv. 7 (85) (2017)
54136–54141.
[21] H. Xu, D. Yin, T. Liu, F. Chen, Y. Chen, X. Wang, J. Sheng, Tea polysaccharide
inhibits RANKL-induced osteoclastogenesis in RAW264.7 cells and ameliorates
ovariectomy-induced osteoporosis in rats, Biomed. Pharmacother. 102 (2018)
539–548.
[22] T. Liu, S. Ding, D. Yin, X. Cuan, C. Xie, H. Xu, X. Wang, J. Sheng, Pu-erh tea extract
ameliorates ovariectomy-induced osteoporosis in rats and suppresses osteoclasto-
genesis in vitro, Front. Pharmacol. 8 (2017) 324.
[23] H. Xu, Y. Wang, Y. Chen, P. Zhang, Y. Zhao, Y. Huang, X. Wang, J. Sheng,
Subcellular localization of galloylated catechins in tea plants [Camellia sinensis (L.)
O. Kuntze] assessed via immunohistochemistry, Front. Plant Sci. 7 (2016) 728.
[24] Y. Takino, H. Imagawa, Y. Aoki, T. Ozawa, Studies on the mechanism of the oxi-
dation of tea leaf catechins, Agric. Chem. Soc. Jpn. J. 45 (8) (1971) 176–183.
[25] S. Takuya, M. Makoto, M. Yosuke, T. Takashi, K. Isao, Biomimetic one-pot pre-
paration of a black tea polyphenol theasinensin A from epigallocatechin gallate by
treatment with copper(II) chloride and ascorbic acid, Chem. Pharm. Bull. 59 (9)
(2011) 1183–1185.
Declaration of Competing Interest
The authors declare no conflict of interest.
References
[26] T. Hatano, M. Hori, M. Kusuda, T. Ohyabu, H. Ito, T. Yoshida, Characterization of
the oxidation products of (-)-epigallocatechin gallate, a bioactive tea polyphenol, on
incubation in neutral solution, Heterocycles 63 (2004) 1547–1554.
[27] L. Xie, Y. Guo, C. Bo, Y. Jing, Epimerization of epigallocatechin gallate to gallo-
catechin gallate and its anti-diabetic activity, Med. Chem. Res. 22 (7) (2013)
3372–3378.
[28] C.L. Shen, J.K. Yeh, J.J. Cao, J.S. Wang, Green tea and bone metabolism, Nutr. Res.
29 (7) (2009) 437–456.
[29] Y. Oka, S. Iwai, H. Amano, Y. Irie, K. Yatomi, K. Ryu, S. Yamada, K. Inagaki,
K. Oguchi, Tea polyphenols inhibit rat osteoclast formation and differentiation, J.
Pharmacol. Sci. 118 (1) (2012) 55–64.
[30] C.H. Ko, K.M. Lau, W.Y. Choy, P.C. Leung, Effects of tea catechins, epigallocatechin,
gallocatechin, and gallocatechin gallate, on bone metabolism, J. Agric. Food Chem.
57 (16) (2009) 7293–7297.
[31] J. Xi, Q. Li, X. Luo, J. Li, L. Guo, H. Xue, G. Wu, Epigallocatechin3gallate protects
against secondary osteoporosis in a mouse model via the Wnt/betacatenin signaling
pathway, Mol. Med. Rep. 18 (5) (2018) 4555–4562.
[1] W.J. Boyle, W.S. Simonet, D.L. Lacey, Osteoclast differentiation and activation,
Nature 423 (6937) (2003) 337–342.
[2] B. Kim, K.Y. Lee, B. Park, Icariin abrogates osteoclast formation through the reg-
ulation of the RANKL-mediated TRAF6/NF-kappaB/ERK signaling pathway in
Raw264.7 cells, Phytomedicine 51 (2018) 181–190.
[3] T. Negishi-Koga, H. Takayanagi, Ca2+-NFATc1 signaling is an essential axis of
osteoclast differentiation, Immunol. Rev. 231 (1) (2009) 241–256.
[4] B. Langdahl, S. Ferrari, D.W. Dempster, Bone modeling and remodeling: potential as
therapeutic targets for the treatment of osteoporosis, Ther. Adv. Musculoskelet. Dis.
8 (6) (2016) 225–235.
[5] T. Liu, Z. Xiang, F. Chen, D. Yin, Y. Huang, J. Xu, L. Hu, H. Xu, X. Wang, J. Sheng,
Theabrownin suppresses in vitro osteoclastogenesis and prevents bone loss in
ovariectomized rats, Biomed. Pharmacother. 106 (2018) 1339–1347.
[6] L. Zhou, Q. Liu, G. Hong, F. Song, J. Zhao, J. Yuan, J. Xu, R.X. Tan, J. Tickner,
Q. Gu, Cumambrin A prevents OVX-induced osteoporosis via the inhibition of os-
teoclastogenesis, bone resorption, and RANKL signaling pathways, FASEB J 33 (6)
(2019) 6726–6735.
[32] T. Tominari, R. Ichimaru, S. Yoshinouchi, C. Matsumoto, K. Watanabe, M. Hirata,
F.M.W. Grundler, M. Inada, C. Miyaura, Effects of O-methylated (-)-epigalloca-
techin gallate (EGCG) on LPS-induced osteoclastogenesis, bone resorption, and al-
veolar bone loss in mice, FEBS Open Bio 7 (12) (2017) 1972–1981.
[33] J.H. Lee, H. Jin, H.E. Shim, H.N. Kim, H. Ha, Z.H. Lee, Epigallocatechin-3-gallate
inhibits osteoclastogenesis by down-regulating c-Fos expression and suppressing the
nuclear factor-kappaB signal, Mol. Pharmacol. 77 (1) (2010) 17–25.
[34] K. Kim, T.H. Kim, H.J. Ihn, J.E. Kim, J.Y. Choi, H.I. Shin, E.K. Park, Inhibitory effect
of purpurogallin on osteoclast differentiation in vitro through the downregulation
of c-Fos and NFATc1, Int. J. Mol. Sci. 19 (2) (2018).
[35] C. Liu, X. Chen, X. Zhi, W. Weng, Q. Li, X. Li, Y. Zou, J. Su, H.G. Hu, Structure-based
development of an osteoprotegerin-like glycopeptide that blocks RANKL/RANK
interactions and reduces ovariectomy-induced bone loss in mice, Eur. J. Med.
Chem. 145 (2018) 661–672.
[36] S.Y. Lin, L. Kang, C.Z. Wang, H.H. Huang, T.L. Cheng, H.T. Huang, M.J. Lee,
Y.S. Lin, M.L. Ho, G.J. Wang, C.H. Chen, (-)-Epigallocatechin-3-gallate (EGCG)
enhances osteogenic differentiation of human bone marrow mesenchymal stem
cells, Molecules 23 (12) (2018).
[7] V. Nagy, J.M. Penninger, The RANKL-RANK story, Gerontology 61 (6) (2015)
534–542.
[8] K. Chen, P. Qiu, Y. Yuan, L. Zheng, J. He, C. Wang, Q. Guo, J. Kenny, Q. Liu,
J. Zhao, J. Chen, J. Tickner, S. Fan, X. Lin, J. Xu, Pseurotin a inhibits osteoclasto-
genesis and prevents ovariectomized-induced bone loss by suppressing reactive
oxygen species, Theranostics 9 (6) (2019) 1634–1650.
[9] K. Zhao, Y. Jia, J. Peng, C. Pang, T. Zhang, W. Han, J. Jiang, X. Lu, J. Zhu, Y. Qian,
Anacardic acid inhibits RANKL-induced osteoclastogenesis in vitro and prevents
ovariectomy-induced bone loss in vivo, FASEB J. 33 (8) (2019) 9100–9115.
[10] Z.M. Chen, Z. Lin, Tea and human health: biomedical functions of tea active com-
ponents and current issues, J. Zhejiang Univ. Sci. B 16 (2) (2015) 87–102.
[11] H. Xu, T. Liu, L. Hu, J. Li, C. Gan, J. Xu, F. Chen, Z. Xiang, X. Wang, J. Sheng, Effect
of caffeine on ovariectomy-induced osteoporosis in rats, Biomed. Pharmacother.
112 (2019) 108650.
[12] L.A. Nash, W.E. Ward, Tea and bone health: findings from human studies, potential
mechanisms, and identification of knowledge gaps, Crit. Rev. Food Sci. Nutr. 57 (8)
(2017) 1603–1617.
[13] N. Khan, H. Mukhtar, Tea polyphenols in promotion of human health, Nutrients 11
9