Synthesis and Biological Evaluation of Celastrol Derivatives with Improved Cytotoxic Selectivity and Antitumor Activities

Article abstract of DOI:10.1021/acs.jnatprod.1c00262

Cdc37 associates kinase clients to Hsp90 and promotes the development of cancers. Celastrol, a natural friedelane triterpenoid, can disrupt the Hsp90-Cdc37 interaction to provide antitumor effects. In this study, 31 new celastrol derivatives, 2a - 2d , 3a - 3g , and 4a - 4t , were designed and synthesized, and their Hsp90-Cdc37 disruption activities and antiproliferative activities against cancer cells were evaluated. Among these compounds, 4f , with the highest tumor cell selectivity (15.4-fold), potent Hsp90-Cdc37 disruption activity (IC₅₀= 1.9 μM), and antiproliferative activity against MDA-MB-231 cells (IC₅₀= 0.2 μM), was selected as the lead compound. Further studies demonstrated 4f has strong antitumor activities both in vitro and in vivo through disrupting the Hsp90-Cdc37 interaction and inhibiting angiogenesis. In addition, 4f exhibited less toxicity than celastrol and showed a good pharmacokinetics profile in vivo. These findings suggest that 4f may be a promising candidate for development of new cancer therapies.

Full text of DOI:10.1021/acs.jnatprod.1c00262

pubs.acs.org/jnp
Article
Synthesis and Biological Evaluation of Celastrol Derivatives with
Improved Cytotoxic Selectivity and Antitumor Activities
Xiao-Long Hu,# Qi-Wei He,# Huan Long, Li-Xin Zhang, Rong Wang, Bao-Lin Wang, Jia-Hao Feng,
Quan Wang, Ji-Qin Hou, Xiao-Qi Zhang, Wen-Cai Ye, and Hao Wang*
Cite This: J. Nat. Prod. 2021, 84, 1954−1966
Read Online
sı
*
Metrics & More
Article Recommendations
Supporting Information
ACCESS
ABSTRACT: Cdc37 associates kinase clients to Hsp90 and promotes the development of cancers. Celastrol, a natural friedelane
triterpenoid, can disrupt the Hsp90−Cdc37 interaction to provide antitumor effects. In this study, 31 new celastrol derivatives, 2a−
2d, 3a−3g, and 4a−4t, were designed and synthesized, and their Hsp90−Cdc37 disruption activities and antiproliferative activities
against cancer cells were evaluated. Among these compounds, 4f, with the highest tumor cell selectivity (15.4-fold), potent Hsp90−
Cdc37 disruption activity (IC50 = 1.9 μM), and antiproliferative activity against MDA-MB-231 cells (IC50 = 0.2 μM), was selected as
the lead compound. Further studies demonstrated 4f has strong antitumor activities both in vitro and in vivo through disrupting the
Hsp90−Cdc37 interaction and inhibiting angiogenesis. In addition, 4f exhibited less toxicity than celastrol and showed a good
pharmacokinetics profile in vivo. These findings suggest that 4f may be a promising candidate for development of new cancer
therapies.
elastrol, a quinone methide friedelane-type triterpene, is
an active constituent derived from the roots of
derivatives incorporated with hydrophobic structures at the C-
29 position can promote disruptions to Hsp90−Cdc37
interactions (Figure 1).7,8 Recently, our group has developed
three types of celastrol derivatives, namely, C-6-indole-
substituted, C-29-esterified, and C-29-amidated celastrol
derivatives, and demonstrated their immunosuppressive effects
in vitro.12 Among these compounds, 1a−1o, in which the C-29
position is substituted by hydrophobic units (Table 1), may
have more potent Hsp90−Cdc37 disruption activities and
antiproliferative activities than celastrol.
In this study, a hit compound was developed from the
celastrol derivatives 1a−1o. On this basis, “hit-to-lead”
optimization resulted in 31 new celastrol derivatives, and
their Hsp90−Cdc37 disruption activities and antiproliferative
C
Tripterygium wilfordii Hook f.1 Celastrol exerts potent
antiproliferative effects on several cancer cell lines, such as
gliomas, hepatocellular carcinomas, prostate cancer, and breast
cancer.2 Although celastrol has potent cytotoxic effects, its high
toxicity results in poor selectivity for tumor cells versus
nonmalignant cells, severely restricting its clinical applica-
tion.3,4 In recent years, efforts have been devoted to the
optimization of celastrol through structural modification,
leading to the production of several celastrol derivatives with
enhanced biological effects; however, their toxicity has not
been improved.5−8 Reduced toxicity and enhanced efficacy
remain the key targets for celastrol modification.
Heat shock protein90 (Hsp90, 90 kDa), which is overex-
pressed in cancer cells, is a promising anticancer target.9 Hsp90
forms a complex with cell division cycle37 (Cdc37), which
targets several protein kinases related to cancer occurrence and
development.10 The disruption of Hsp90−Cdc37 interactions
can inhibit the proliferation of cancer cells.10 Evidence suggests
that celastrol, an Hsp90 inhibitor, exerts a potential antitumor
effect by disrupting Hsp90−Cdc37 complexes.11 Celastrol
Received: March 19, 2021
Published: June 25, 2021
© 2021 American Chemical Society and
American Society of Pharmacognosy
https://doi.org/10.1021/acs.jnatprod.1c00262
J. Nat. Prod. 2021, 84, 1954−1966
1954

Journal of Natural Products
pubs.acs.org/jnp
Article
caspase-3-activating effects also show strong antiproliferative
activities against nonmalignant cells (Table 3). Hence, it was
speculated that their cytotoxicity to nonmalignant cells
(hMSC) was related to their effects on caspase-3. Docking
analysis showed that the A ring of celastrol and its derivatives,
especially at the C-3 position, can form several key H-bond
interactions with caspase-3, which lead to their caspase-3-
activating effects.13 Therefore, introducing groups that can
weaken the interaction between celastrol and caspase-3 at the
C-3 position may be an efficient strategy for toxicity reduction.
Some preliminary modifications were conducted to afford
three series of derivatives of 1i, namely, 2a−2d, 3a−3g, and
4a−4d. The antiproliferative effects against nonmalignant cells
(hMSCs) and the caspase-3-activated effects of the celastrol
derivatives decreased significantly compared with those of 1i
(Table 4). Notably, 4a−4d, with carbamate groups substituted
at the C-3 position of 1i, showed selectivity that was over 10
times that of 1i (Table 4). Therefore, another 16 1i derivatives
(4e−4t), in which the C-3 position was substituted by
carbamate groups, were synthesized. As a result, 4e−4t
exhibited relatively weak activated effects on caspase-3 (IC50
Figure 1. Summary of celastrol positions of structural modification.
activities were evaluated. Furthermore, the in vitro and in vivo
antitumor effects, mechanisms of action, toxicity, and
preliminary pharmacokinetics of the lead compound 4f were
investigated.
RESULTS AND DISCUSSION
■
The celastrol derivatives 1a−1o (Table 1) were synthesized
through nucleophilic reactions between celastrol and organic
halides in N,N-dimethylformamide at room temperature.
NaHCO3 was used as the base (see previously published
work for details12). Then, the C-3 ester derivatives 2a−2d
were synthesized through the reaction of compound 1i with
the corresponding anhydride or acyl chloride (Scheme 1).
Next, the C-3 α,β-unsaturated ester derivatives, 3a−3g, were
synthesized (Scheme 2). Finally, a series of C-3 carbamate-
substituted celastrol derivatives, 4a−4t, were synthesized
through the reaction of compound 1i with the corresponding
carbamoyl chloride (Scheme 3).
values ranging from 9.8 to 24.3 μM), compared with 1i (IC50
=
1.78 μM) and showed relatively poor toxic effects on
nonmalignant cells (IC50 values ranging from 0.9 to 2.8 μM),
compared with 1i (IC50 = 0.8 μM; Table 5). Accordingly, the
introduction of a carbamate group with large hindrance and
rigidity at the C-3 position weakened the activating effects of 1i
on caspase-3. The compound 4f, which had the highest
selectivity (15.4-fold) for tumor MDA-MB-231 cells versus
nonmalignant cells (hMSCs) and potent antiproliferative
activity, was selected as the lead compound. Figure 2 shows
that C-3 carbamates afforded 10-fold reduction in caspase-3-
activating potency while maintaining Hsp90−Cdc37 disrupting
potency.
Docking results further supported the observation that
introducing a carbamate group to the C-3 position of 1i can
reduce the binding of 1i to caspase-3 (Figure 3). The
characteristics of the bonds between the test compounds (1i
and 4f) and caspase-3 were evaluated with thermodynamic and
kinetic experiments. In the thermodynamic experiments, a
cellular thermal shift assay was used in studying the thermal
stabilization of proteins upon ligand binding.14,15 This method
has been used extensively on purified proteins for the detection
of interactions between donors and ligands. The thermal
stabilization of a protein can be stabilized by ligands that bind
to it. As shown in Figure 4A and B, the levels of caspase-3
decreased with increasing temperature in the DMSO-treated
group. However, treatment with 1i and 4f increased the
tolerance temperature of caspase-3, and the effect of 1i was
greater. In kinetic experiments, surface plasmon resonance
(SPR) was used in evaluating the binding affinity between the
test compounds and caspase-3. SPR is the most recognized
method for studying dynamic properties (such as the
association rate constant, dissociation rate constant, and
affinity) between ligands and donors.16 As shown in Figure
4C and D, a stronger binding affinity was found between 1i
and caspase-3, with a KD value of 0.9 μM, compared with that
between caspase-3 and 4f (KD value = 3.8 μM). These results
demonstrated that introducing carbamate groups at the C-3
position can reduce the binding between celastrol derivatives
and caspase-3 and thereby decrease the activated effects of
celastrol derivatives on caspase-3 and toxicity.
The Hsp90−Cdc37 disruption activities and antiproliferative
activities of celastrol derivatives (1a−1o) were evaluated.
Table 1 shows that compounds 1f, 1i, and 1l−1o at a 50 μM
concentration exhibited Hsp90−Cdc37 disruption rates of
more than 90%, which was higher than that of celastrol
(82.3%). Thus, the IC50 values of the compounds were further
evaluated. The results showed that 1i exhibited the most
potent Hsp90−Cdc37 disruption activity, with an IC50 value of
1.7 μM (Figure 2). Table 2 shows that six compounds
exhibited potent antiproliferative activities against A549,
MDA-MB-231, KB, and MCF-7 cells, especially 1i against
MDA-MB-231 cells (IC50 = 0.2 μM). These compounds with
potent Hsp90−Cdc37 disruption activities also exhibited
strong antiproliferative activities. However, these derivatives
also exhibited strong cytotoxicity to nonmalignant cell lines
(human mesenchymal bone marrow stem cells, or hMSCs),
with IC50 values ranging from 0.6 to 1.5 μM. Thus, they
showed poor selectivity for tumor cells versus nonmalignant
cells (less than 4-fold; Table 3). Therefore, further structure
modifications focusing on selectivity enhancement were
necessary, and 1i, the most potent analogue, was selected as
a hit compound for structural optimization.
Celastrol and celastrol derivatives have been found to exhibit
effects on caspase-3 and thereby induce the apoptosis of cancer
cells.13 Caspase-3, as a key protein in the mitochondrial
apoptotic pathway, strongly inhibits the proliferation of tumors
and normal cells in a nonselective manner.14,15 As shown in
Table 3, compounds 1f, 1i, and 1l−1o exhibited potent
caspase-3-activating effects, with EC50 values ranging from 1.2
to 4.4 μM (Table 3). Notably, compounds with potent
1955
https://doi.org/10.1021/acs.jnatprod.1c00262
J. Nat. Prod. 2021, 84, 1954−1966

Journal of Natural Products
pubs.acs.org/jnp
Article
Table 1. Chemical Structures and Hsp90−Cdc37 Disruption Activities of Compounds 1a−1o and Celastrol
Previous studies have demonstrated that celastrol can
disrupt the interaction between the N-terminal domain of
Hsp90 (Hsp90N) and the middle domain of Cdc37 (Cdc37M).
In the structure of the Hsp90N−Cdc37M complex, an
important H-bond between Q133 (Hsp90N) and R166/R167
(Cdc37M) was identified as the key interaction for the activity
of the Hsp90N−Cdc37M complex.11 In the present inves-
tigation, the interaction between 4f and the hydrophobic
pocket of the human Hsp90N−Cdc37M complex (PDB code:
2K5B) was mimicked through molecular docking, and the
mechanism of 4f in disrupting the complex was explained. As
shown in Figure 5A and B, 4f was docked with the
hydrophobic pocket of the Hsp90N−Cdc37M complex and
formed four H-bonds with Trp168, Arg166, Arg167, and
Lys242 of the Cdc37 protein (purple). These H-bonds might
allow 4f to efficiently disrupt the key interaction (H-bonds
between Q133Hsp90 and R166/R167Cdc37) of the Hsp90N−
Cdc37M complex by binding to Cdc37 (noncovalent bond),
thereby affecting the stability of the Hsp90−Cdc37 complex.
An immunoprecipitation experiment was carried out, and
DMSO was used as the negative control. The purpose was to
verify whether 4f disrupted the interaction of Hsp90−Cdc37 in
vitro. As shown in Figure 6, no nonspecific proteins affected
this procedure. Compound 4f led to a decrease in the amount
of the Cdc37 protein compared with celastrol when Hsp90α
was pulled down. The data obtained suggested that 4f
1956
https://doi.org/10.1021/acs.jnatprod.1c00262
J. Nat. Prod. 2021, 84, 1954−1966

Journal of Natural Products
pubs.acs.org/jnp
Article
a
Scheme 1. Synthetic Route for the Target Compounds 2a−2d
a
Reagents and conditions: (a) Et3N, CH2Cl2, 0 °C to rt 12 h for 2a and 2b; (b) Et3N, CH2Cl2, −10 °C to rt, 12 h for 2c and 2d.
a
Scheme 2. Synthetic Route for the Target Compounds 3a−3g
a
Reagents and conditions: (a) Et3N, CH2Cl2, −10 °C to rt, 12 h.
1957
https://doi.org/10.1021/acs.jnatprod.1c00262
J. Nat. Prod. 2021, 84, 1954−1966

Journal of Natural Products
pubs.acs.org/jnp
Article
a
Scheme 3. Synthetic Route for the Target Compounds 4a−4t
a
Reagents and conditions: (a) Et3N, DMAP, CH2Cl2, 0 °C to rt, 12 h.
Evidence has shown that disruption of the Hsp90−Cdc37
complex will downregulate the expression of Hsp90−Cdc37
clients, resulting in the activation of the mitochondrial
apoptotic pathway.17 In this regard, the effects of 4f and
celastrol on the expression of Hsp90−Cdc37 clients (such as
Cdk4 and p-Akt) and mitochondria-related apoptotic proteins
in MDA-MB-231 cells were detected. As shown in Figure 7A
and B, compound 4f (100 and 200 nM) downregulated the
expression of p-Akt and Cdk4 in a dose-dependent manner. In
addition, the ratio of Bax/Bcl-2, the release of cytochrome c,
and the activities of caspase-3 and -9 (Figure 7C) were
increased by 4f. The activity of celastrol (200 nM) was
significantly weaker than that of 4f. These results suggested
that 4f inhibited the expression of Hsp90−Cdc37 client
proteins.
Figure 2. Hsp90−Cdc37 interaction disruption activities of selected
celastrol derivatives. Sigmoidal dose−response curves and IC50 values
are presented as means SD (n = 3).
Hsp90−Cdc37 disruption can induce cell cycle arrest and
cell apoptosis and then exert antitumor activity.18 The results
showed that 4f arrested the cell cycle at the G0/G1 phase in a
possesses stronger disruption effects on the Hsp90−Cdc37
complex when compared with celastrol at the cellular level.
1958
https://doi.org/10.1021/acs.jnatprod.1c00262
J. Nat. Prod. 2021, 84, 1954−1966

Journal of Natural Products
pubs.acs.org/jnp
Article
Table 2. Antiproliferative Activities of Compounds 1a−1o
and Celastrol against the A549, MDA-MB-231, KB, and
MCF-7 Cancer Cell Lines
Table 4. Antiproliferative Effects, Caspase-3-Activating
Effects, and Selectivity of Compounds 2a−2d, 3a−3g, and
4a−4d
a
cell line/IC50 value (μM)
caspase-3
hMSCs
MDA-MB-
activation (EC50,
(IC50,
231 (IC ,
b50
a
b
c
compound
A549
MDA-MB-231
KB
MCF-7
compound
μM)
μM)
μM)
selectivity
1a
1b
1c
1d
1e
1f
1g
1h
1i
1j
1k
1.1 0.1
1.7 0.0
1.1 0.0
1.3 0.1
4.2 0.4
1.0 0.2
1.0 0.0
0.8 0.1
0.8 0.0
1.6 0.0
6.3 0.3
1.0 0.0
1.3 0.1
1.1 0.1
1.1 0.2
1.8 0.1
0.8 0.0
0.7 0.0
0.8 0.0
0.9 0.1
1.1 0.1
0.6 0.0
0.4 0.1
0.8 0.0
0.2 0.0
0.1 0.1
0.8 0.1
0.4 0.0
0.4 0.0
0.5 0.1
0.6 0.1
0.9 0.1
0.9 0.0
1.0 0.1
0.6 0.1
0.6 0.0
1.9 0.1
0.8 0.0
0.6 0.0
0.3 0.0
0.4 0.0
1.0 0.2
1.9 0.2
0.7 0.0
0.6 0.1
0.8 0.1
0.7 0.0
1.7 0.1
0.9 0.1
1.0 0.1
0.5 0.1
0.6 0.1
2.0 0.1
0.9 0.0
0.60 0.05
0.4 0.1
0.4 0.1
1.1 0.1
1.8 0.3
0.7 0.1
0.7 0.1
1.0 0.3
0.8 0.1
1.2 0.1
2a
2b
2c
2d
3a
3b
3c
3d
3e
3f
3g
4a
4b
4c
4d
19.3 2.5
20.9 2.0
16.4 1.5
21.2 3.8
22.1 3.3
19.7 2.9
20.1 3.0
14.1 1.5
15.7 1.9
17.8 2.9
14.9 1.2
20.4 2.0
18.7 3.2
19.9 1.4
18.5 2.0
2.4 0.3
2.6 0.9
2.2 0.1
2.0 0.2
3.7 0.4
3.0 0.3
2.8 0.3
1.4 0.2
2.0 0.2
2.9 0.2
2.4 0.3
3.2 0.5
3.0 0.3
3.1 0.2
2.9 0.6
0.5 0.1
0.4 0.
5.4
6.7
7.5
5.5
5.5
6.1
6.0
5.5
6.8
7.1
5.9
12.3
10.2
13.3
0.3 0.0
0.4 0.1
0.7 0.1
0.5 0.0
0.5 0.1
0.3 0.0
0.3 0.0
0.4 0.1
0.4 0.0
0.3 0.0
0.3 0.0
0.2 0.0
0.2 0.0
1l
1m
1n
1o
celastrol
13.5
b
a
EC50 values were determined using a caspase-3 activity assay. IC50
a
IC50 values were determined by an MTT assay after 48 h long
incubations with different concentrations of each compound, and the
data are expressed as the means SD (n = 6).
values were determined using an MTT assay after 48 h incubations
with different concentrations of each compound. Selectivity refers to
the ratio of IC50 value of hMSCs cells to MDA-MB-231 cells. Data are
c
expressed as the means SD (n = 6).
Table 3. Selectivity and Caspase-3 Activating Effects of
Compounds 1f, 1i, 1l−1o, and Celastrol
Table 5. Antiproliferative Effects, Caspase-3-Activating
Effects, and Selectivity of 4e−4t
hMSCs (IC50,
caspase-3 activation (EC50,
a
b
c
compound
μM)
μM)
selectivity
caspase-3
hMSCs
MBA-MD-231
activation (EC50,
(IC50,
cells (IC ,
b50
1f
1i
1l
1m
1n
1.5 0.4
0.8 0.1
0.9 0.2
1.0 0.4
0.6 0.0
0.7 0.1
0.7 0.1
4.4 1.6
1.8 0.2
2.1 0.2
1.6 0.1
1.1 0.1
1.7 0.2
1.4 0.1
2.6
4.0
2.5
2.5
1.2
1.2
0.8
a
b
c
compound
μM)
μM)
μM)
selectivity
4e
4f
4g
4h
4i
4j
4k
4l
4m
4n
4o
4p
4q
4r
4s
4t
23.9 2.1
25.5 1.7
10.0 1.0
24.1 2.0
24.3 2.2
23.1 1.1
20.5 2.0
17.5 0.9
20.5 1.1
20.3 1.4
22.3 2.0
19.3 1.0
20.1 1.2
20.3 1.5
19.6 1.7
18.2 1.4
2.0 0.1
2.8 0.2
1.0 0.1
2.4 0.2
2.7 0.3
2.2 0.2
2.0 0.1
0.9 0.1
0.9 0.0
1.0 0.0
1.2 0.2
1.8 0.1
1.6 0.1
1.9 0.0
1.8 0.1
1.8 0.4
0.2 0.0
0.2 0.1
0.3 0.0
0.1 0.0
0.2 0.0
0.2 0.0
0.3 0.0
0.3 0.0
0.3 0.1
0.3 0.0
0.1 0.0
0.2 0.0
0.2 0.0
0.2 0.0
0.3 0.0
0.3 0.1
11.8
15.4
2.9
12.2
13.9
13.6
6.8
2.7
3.0
3.9
8.4
8.5
8.3
10.1
7.2
6.5
1o
celastrol
a
IC50 values were determined using an MTT assay after a 48 h long
b
incubation with different concentrations of each compound. EC50
values were determined using a caspase-3 activity assay. Selectivity
c
refers to the ratio of IC50 value of hMSCs cells to MDA-MB-231 cells.
Data are expressed as the means SD (n = 6).
dose-dependent manner (Figure S1, Supporting Information),
suggesting that the cell cycle arrest might be one of the
mechanisms of its antiproliferative effects. In addition, the
results showed also that 4f could significantly damage the cells
in a dose-dependent manner, and 4f exhibited a more potent
cytotoxic effect than celastrol at a concentration of 200 nM
(Figure S2, Supporting Information). These results suggested
that the antiproliferative effect of 4f was involved in cell cycle
arrest and mitochondrial apoptosis.
a
b
EC50 values were determined using a caspase-3 activity assay. IC50
values were determined using an MTT assay after 48 h incubations
c
with different concentrations of each compound. Selectivity refers to
the ratio of IC50 value of HMSCs cells to MDA-MB-231 cells. Data
are expressed as the means SD (n = 6).
The antitumor effect of 4f in vivo was determined with a
xenograft model generated by grafting MDA-MB-231 cells in
nude mice (Figure 8A). Figures 9B and 8C show that 4f
inhibited tumor growth at an oral dose of 10 mg/kg, showing a
greater potency than either celastrol or the positive control,
carboplatin (CBP). Western blot analysis demonstrated that 4f
can inhibit the expression of Hsp90−Cdc37 clients and
activate mitochondrial-related apoptotic proteins in vivo
(Figure 8E,F), indicating that the antitumor effects of 4f in
vivo were related also to Hsp90−Cdc37 disruption.
Compound 4f showed excellent in vivo preliminary
pharmacokinetic properties in rats that received the compound
at a dose of 10 mg/kg (Figure S3, Supporting Information).
Furthermore, to prove that 4f is less toxic than celastrol in vivo,
the body weights of xenografted mice and tested blood toxicity
were monitored, and visceral HE staining was conducted on
the last day of the experiment. The results showed that
celastrol and 4f can reduce the body weights of mice as the
time of administration increased, but the negative effect of 4f
1959
https://doi.org/10.1021/acs.jnatprod.1c00262
J. Nat. Prod. 2021, 84, 1954−1966

Journal of Natural Products
pubs.acs.org/jnp
Article
Figure 3. Possible binding modes of 1i and 4f toward caspase-3 (PDB code: 1GFW). (A) Chemical structure of 1i and the binding mode of 1i and
caspase-3. (B) Chemical structure of 4f and the binding mode of 4f and caspase-3. The green dotted line represents H-bond interactions, and the
purple dotted line represents π−π interactions.
Figure 4. Binding characters of compounds 1i and 4f with caspase-3. (A) Original bands of CESTA, which were used to evaluate the binding
between the test compounds (1i and 4f) and caspase-3 at the thermodynamic levels. (B) Quantitative analysis of CESTA. (C) SPR was used to
assess the binding between 1i and caspase-3 at the kinetic level. (D) SPR was used to assess the binding between 4f and caspase-3 at the kinetic
level.
on these weights was significantly less than that of celastrol
(Figure 9A). When compared with a vehicle control group,
celastrol significantly reduced the number of white blood cells
(Figure 9B), red blood cells (Figure 9C), and platelets (Figure
9D), and no significant difference was found between 4f and
the vehicle group (p > 0.05). HE staining showed that celastrol
can cause damage to the heart, liver, lungs, and kidneys, but
the degree of damage by 4f to these organs was much lower
than that caused by celastrol (Figure 9E). These results
demonstrated that 4f appeared to be less toxic than celastrol in
vivo.
Carbamate groups can inhibit angiogenesis and promote
antiproliferation, thereby exerting antitumor effects.19 To
confirm whether 4f can inhibit angiogenesis, an in vitro
wound-scratch assay and an in vivo zebrafish model were used.
Figure S4 (Supporting Information) shows that after treatment
with the test compounds at different intervals (0, 12, and 24
h), cells in the celastrol (0.1 μM) and 1i (0.05 μM) groups
moved fast into the scraped area over time. Treatment with 4f
(0.05 μM) resulted in a significant decrease in migration
ability. The quantitative results were determined as a migration
ratio, which indicated differences in migration ratio to 0 h
1960
https://doi.org/10.1021/acs.jnatprod.1c00262
J. Nat. Prod. 2021, 84, 1954−1966

Journal of Natural Products
pubs.acs.org/jnp
Article
Figure 5. (A) Molecular docking of 4f to Hsp90N−Cdc37M. (B) Possible binding mode of 4f with the Hsp90N−Cdc37M complex.
Figure 6. Compound 4f disrupted the Hsp90−Cdc37 interaction in MDA-MB-231 cells. (A) Expression levels of Cdc37 and Hsp90α after Hsp90α
were pulled down; IgG is to exclude interference from nonspecific proteins. (B) Grayscale analysis is presented by means of the density ratios of
proteins treated with 4f or celastrol to those untreated. **p < 0.01 compared with the celastrol-treated group.
Figure 7. Effects of 4f and celastrol on Hsp90 client proteins and apoptosis-related proteins in MDA-MB-231 cells. (A) The original bands of the
expression levels of Hsp90, Cdk4, p-Akt, Bax, and Bcl-2, β-actin were used as a loading control. (B) Grayscale analysis was conducted by means of
the density ratios of the proteins to β-actin. (C) Activities of caspase-3 and caspase-9. *p < 0.05, ** p < 0.01, or *** p < 0.001 compared with
control group.
groups. Hence, 2f inhibited the migration of human umbilical
vein endothelial cells (HUVECs) in a time-dependent manner.
Zebrafish have been proposed as a valid alternative animal
model for investigating angiogenesis, and the effects of
celastrol, 1i, and 4f on embryonic angiogenesis in zebrafish
were evaluated. Figure 10 shows the results of treating
zebrafish embryos with celastrol, 1i, 4f, or the vehicle control.
Treating live zebrafish embryos with different concentrations
of 4f (0.05, 0.1, and 0.2 μM) blocked the formation of
intersegmental vessels while preserving fluorescence in the
dorsal aorta and major cranial vessels in a dose-dependent
manner. However, celastrol and 1i showed no obvious
inhibitory effects on angiogenesis. Overall, these results
indicated that 4f exhibited antiangiogenic effects in vitro and
1961
https://doi.org/10.1021/acs.jnatprod.1c00262
J. Nat. Prod. 2021, 84, 1954−1966

Journal of Natural Products
pubs.acs.org/jnp
Article
Figure 8. Antitumor effects of 4f and celastrol in vivo. (A) Schematic of the in vivo experimental procedure. (B) Tumors from nude mice at the
final day. (C) Tumor volumes of MDA-MB-231 xenografts in each treatment group (n = 6). (D) Original bands of Hsp90 client proteins and Bax
and Bcl-2 in tumor tissues (n = 6). (E) Grayscale analysis of bands. (F) Caspase-3 and caspase-9 activities in tumor tissues (n = 6). *p < 0.05, **p <
0.01, or ***p < 0.001 compared with the control group.
acyl chloride, which was used in subsequent reactions without further
purification. A solution of compound li (0.2 mmol, 1.0 equiv) and
TEA (0.4 mmol) in dry dichloromethane was added dropwise to the
corresponding aforementioned acyl chloride intermediate (0.3 mmol,
1.5 equiv) at −10 °C under a nitrogen atmosphere. After stirring at 25
°C for 12 h, the solvent was removed under reduced pressure. The
residue was purified further by flash column chromatography on silica
gel (hexanes/EtOAc) to produce 2c and 2d as yellow powders.
General Procedures for the Synthesis of Compounds 3a−
3g. Oxalyl chloride (5.7 mmol, 1.5 equiv) and DMF (6 drops) were
added to the corresponding acid (3.8 mmol, 1.0 equiv) in dry
dichloromethane (10 mL) at 0 °C. Each mixture was warmed slowly
to room temperature and stirred for 12 h, and then the solvent was
evaporated under vacuum to obtain the corresponding acyl chloride,
which was used in subsequent reactions without further purification.
A solution of li (0.2 mmol, 1.0 equiv) and TEA (0.4 mmol, 2.0 equiv)
in dry dichloromethane was added dropwise to the corresponding
aforementioned acyl chloride intermediate (0.3 mmol, 1.5 equiv) at
−10 °C under a nitrogen atmosphere. After stirring at 25 °C for 12 h,
the solvent was removed under reduced pressure. The residue was
further purified by flash column chromatography on silica gel
(hexanes/EtOAc) to produce 3a−3g as yellow powders.
General Procedures for the Synthesis of Compounds 4a−
4t. Triphosgene (2.5 mmol, 1.0 equiv) and NaHCO3 (7.5 mmol, 3.0
equiv) in dry dichloromethane (5 mL) were added for 30 min at 0 °C
to the corresponding amine (3.8 mmol, 1.5 equiv) in dry
dichloromethane (2.5 mL). Each mixture was warmed slowly to
room temperature and stirred for 12 h, and then the solvent was
evaporated under vacuum to obtain the corresponding carbamoyl
chloride, which was used for subsequent reactions without further
purification. A solution of li (0.2 mmol, 1.0 equiv), TEA (0.4 mmol,
2.0 equiv), and 4-dimethylaminopyridine (DMAP) (1.2 mg, 0.01
mmol, 0.05 equiv) in dry dichloromethane (5 mL) was added
dropwise to the aforementioned carbamoyl chloride intermediate (0.3
mmol, 1.5 equiv) at 0 °C. After stirring at 25 °C for 12 h, the solvent
in vivo, and the introduction of carbamate groups to the C-3
position promoted the antiangiogenetic effects of the celastrol
derivatives.
EXPERIMENTAL SECTION
■
General Experimental Procedures. All reagents were available
from commercial suppliers and were used directly unless otherwise
stated. Melting points were measured on an MP88 melting point
apparatus and were uncorrected. Optical rotations were recorded on
an Anton Paar MCP 150 automatic polarimeter. H NMR and 13C
1
NMR spectra were recorded on a Bruker AV-400/AV-600
spectrometer instrument at 25 °C. High-resolution electrospray
ionization mass spectrometry (HRESIMS) data were acquired on an
ESI-FT-ICR SolariX spectrometer (Bruker). Column chromatography
was conducted with silica gel (200−300 mesh). Routine thin-layer
chromatography (TLC) on silica gel GF254 (Qingdao Haiyang
Chemical Plant, Qingdao, People’s Republic of China) was performed
to monitor the progress of reactions with an ultraviolet lamp. High-
performance liquid chromatography (HPLC) analysis was performed
on a Shimadzu Prominence LC-2030 HPLC instrument with an
Inertsil ODS-3 column (250 mm × 4.6 mm; 5 μm).
General Procedures for the Synthesis of Compounds 2a−
2d. A solution of li (0.2 mmol, 1.0 equiv) and triethylamine (TEA)
(0.4 mmol, 2.0 equiv) in dry dichloromethane (5 mL) was added
dropwise to the corresponding anhydride (0.3 mmol, 1.5 equiv) at 0
°C under a nitrogen atmosphere. After stirring at 25 °C for 10 h,
solvent was removed under reduced pressure. Each residue was
purified further by flash column chromatography on silica gel
(hexanes/EtOAc) to produce 2a and 2b as yellow powders.
Oxalyl chloride (5.7 mmol, 1.5 equiv) and dimethylformamide
(DMF) (6 drops) were added to the corresponding acid (3.8 mmol,
1.0 equiv) in dry dichloromethane (10 mL) at 0 °C. Each mixture was
slowly warmed to room temperature and stirred for 12 h, and then the
solvent was evaporated under vacuum to obtain the corresponding
1962
https://doi.org/10.1021/acs.jnatprod.1c00262
J. Nat. Prod. 2021, 84, 1954−1966

Journal of Natural Products
pubs.acs.org/jnp
Article
Figure 9. Effects of 4f and celastrol on body weight, hemocytes, and histological sections. (A) Body weight from day 16 to day 26. (B) Number of
white blood cells (WBC) on the final day. (C) Number of red blood cells (RBC) on the final day. (D) Number of blood platelets on the final day.
(E) HE staining of the main viscera. *p < 0.05, **p < 0.01, or ***p < 0.001 compared with control group.
was removed under reduced pressure. The residue was further
purified by flash column chromatography on silica gel (hexanes/
EtOAc) to produce 4a−4t as yellow powders.
Full-length active caspase-3, His-tagged full-length Hsp90, GST-
tagged Cdc37, and an immunoprecipitation kit were purchased from
Abcam (Cambridge, MA, USA). A 384-well black low-volume plate
was purchased from Greiner (Germany). Anti-GST-Cryptate
(61GSTKLA) and anti-6His-XL665 (61HISXLA) were purchased
from CISBIO (France). Rabbit antibodies for Hsp90α, Cdc37, Bax,
Bcl-2, p-Akt, and β-actin were purchased from Santa Cruz
Biotechnology (Santa Cruz, CA, USA).
Hsp90−Cdc37 Disruption Activity (HTRF Measurement).
The HTRF experiments were conducted according to a previous
study.20 Briefly, in a 384-well black low-volume plate in PBS buffer
(containing 400 nM KF, pH 7.3), 4 μL of His-tagged full-length
Hsp90 (final concentration: 250 nM) and 4 μL of GST-tagged Cdc37
(final concentration: 250 nM) were mixed into the plate wells for
each assay. Then, 4 μL of the diluted compounds were added and
Biological Materials. All test compounds were synthesized in our
laboratory with a purity over 98%. 3-[4,5-Dimethylthiazol-2-yl]-2,5-
diphenyltetrazolium bromide (MTT), Dulbecco’s modified Eagle’s
medium (DMEM), and carboplatin were purchased from Sigma-
Aldrich (St. Louis, MO, USA). Hoechst 33258 kit, JC-1 staining kit,
PI staining kit, HE staining kit, and annexin V−PI double staining kit
were purchased from KeyGen BioTECH (Nanjing, People’s Republic
of China). RIPA lysis buffer, protein inhibitor, caspase-3 and caspase-
9 activity kits, hematoxylin and eosin (HE) staining kit, and BCA
protein quantification kit were purchased from Beyotime (Shanghai,
People’s Republic of China). Polyvinylidene fluoride (PVDF)
membrane was purchased from Millipore Co. (Billerica, MA, USA).
1963
https://doi.org/10.1021/acs.jnatprod.1c00262
J. Nat. Prod. 2021, 84, 1954−1966

Journal of Natural Products
pubs.acs.org/jnp
Article
Figure 10. Effects of 4f on antiangiogenesis in vivo. Zebrafish embryos treated with either a blank control or test compounds.
mixed. After being incubated at 37 °C for 1 h, 4 μL of anti-GST-
Cryptate (61GSTKLA, CISBIO) and 4 μL of anti-6His-XL665
(61HISXLA, CISBIO) were added and incubated at 37 °C for
another 0.5 h. Then, the time-resolved fluorescence was measured
using a Molecular Devices reader (SpectraMax paradigm; λex = 320
nm, λem = 665 and 620 nm). The HTRF ratio was calculated as a two-
wavelength signal ratio: (signal 665 nm/signal 620 nm) × 10 000.
Cytotoxicity Assays. A549 cells, MDA-MB-231 cells, KB cells,
and MCF-7 cells (American Type Culture Collection, Manassas, VA,
USA) were seeded into 96-well plates (1 × 104 cells/well) and
allowed to culture for 24 h. The cells were then stimulated with
different concentrations of each test compound (0.01−2 μM) for 48
h. Then, an MTT assay was used to detect the cell viability and IC50
values were calculated.21
Surface Plasmon Resonance Assay. The procedures used were
conducted in a previous report.22 Briefly, purified human recombinant
protein caspase-3 (final concentration: 15 μg/mL) was loaded into
the chip at 0.5 μL/s in NaAc buffer (pH 5.5) for 20 min at 4 °C.
Then, different concentrations of each test compounds (10 mM stock
solution in DMSO) in PBS flowed through the surface of the chip at
0.5 μL/s in PBST (pH 7.4) for 600 s at 4 °C. Finally, proteins were
dissociated from the chip at 2 μL/s in glycine-HCl (pH 2.0) for 360 s
at 4 °C. The binding signals (RU) were detected by SPR using a
Biacore T200 (GE Healthcare, Chicago, IL, USA).
Cellular Thermal Shift Assay (CETSA) and Isothermal Dose−
Response Fingerprint (ITDRFCETSA). Cellular thermal shift assay
(CETSA) and isothermal dose−response fingerprint (ITDRF)
experiments were conducted in a similar manner to a previous
report.22 For CESTA experiments, MDA-MB-231 cells were seeded in
10 cm culture dishes, and then the cells were incubated with 1i or 2f
(5 μM) in fresh growth medium for 3 h at 37 °C in a cell incubator.
The same volume of DMSO was used as a negative control in another
10 cm dish. Subsequently, the medium was removed, and the cells
were washed with PBS and distributed into six different microtubes
with 80 μL of cell suspension in each tube for both test compounds
(1i or 2f) and DMSO-treated cells. The microtubes were heated at
designated temperatures (45 to 85 °C) for 8 min on a heating block.
After heating, the microtubes were removed and balanced at 25 °C for
another 3 min. Then, the cells were lysed to extract the proteins and
perform Western blot analysis. For ITDRFCETSA experiments, all
procedures were the same as for the CETSA work, and the cells were
incubated with 1i or 2f in different concentrations at the designated
temperature.
maintained according to standard protocols.19 Compounds were
prepared initially as a 10 mg/mL stock solution in dimethyl sulfoxide
(DMSO). The stock solution was diluted to the relevant assay
concentration with fish water, and 0.5% DMSO served as a vehicle
control. The embryos were distributed in 24-well plates, with 30
embryos placed in each well. Then, 0.05, 0.1, and 0.2 μM of each
compound (CEL, 1i, and 2f) were added. The embryos were exposed
to each compound and incubated at 28.5 °C. After 24 h, the zebrafish
were anesthetized with 0.01% tricaine and imaged under a
fluorescence microscope (Nikon Ts-100, Japan). All studies were
performed in accordance with the approved guidelines from the China
Zebrafish Resource Center, with the approved protocol number
2012CB944504.
In Vivo Xenograft Experiment. Female athymic nude mice (5−
6 weeks old) were used for in vivo xenograft studies. Tumor cell
inoculation procedures were the same as a previous study.21 When
tumors reached an average volume of approximately 100 mm3, mice
were randomized into various treatment groups (n = 6), including
vehicle group, carboplatin group (10 mg/kg), celastrol group (10 mg/
kg), and a group containing 4f (10 mg/kg). On the last day of
administration, a routine blood examination (Mindary, BC-5000 Vet)
and a blood biochemistry examination (Mindary, BS-600) were
conducted. The tumors were processed to perform Western blot
analysis and HE staining. All protocols were approved by the
Experimental Animal Centre of China Pharmaceutical University (no.
SYXK2021-0011, January 25, 2021, to January 24, 2026).
Hematoxylin−Eosin Staining. Histological assessment was
performed as described before.23 Formalin-fixed, paraffin-embedded
tumor tissue was sectioned at 5 mm thickness, and the sections were
stained with hematoxylin and eosin.
Western Blot Analysis. After the drug treatment described above,
cells and tumor tissues were collected and washed two times with
PBS. Protein extraction and Western blot experiments were
conducted in the same manner as a previous work.21 Images were
analyzed using the ImageJ software.
Immunoprecipitation. MDA-MB-231 cells (5 × 105 cells/well)
were seeded into a 6 cm dish and incubated for 24 h. Then, 2f or
celastrol at 200 nM were added into dishes for another 4 h incubation.
The proteins were extracted with lysis buffer. Then, Hsp90α was
pulled down by γ-phosphate-linked ATP-Sepharose, and the Cdc37
was pulled down by Hsp90α-Sepharose. Equal amounts of total
protein were subjected to SDS-PAGE to perform Western blot
analysis.
In Vivo Fluorescent Zebrafish Assay. Transgenic zebrafish (24
hpf, Fli-1) embryos were supplied using the drug screening platform
of the Shandong Academy of Sciences Institute and were grown and
Molecular Docking. The molecular docking study was performed
using Discovery Studio 3.0 (DS3.0) software. Briefly, the ligands (1i
and 2f) and protein were prepared, hydrogen was added, and water
1964
https://doi.org/10.1021/acs.jnatprod.1c00262
J. Nat. Prod. 2021, 84, 1954−1966

Journal of Natural Products
pubs.acs.org/jnp
Article
molecules were deleted by DS Server. The result of docking was
treated with DS4.5 Client. In this study, the crystal structure of
casapse-3 (PDB code: 1GFW) was selected for docking. The ligand
(code: MSI) was employed to define the binding site with xyz
coordinates (37.6468, 33.5653, 27.9733, radius 8.12325 Å). The
protocol, CDOCKER, was used to perform the docking. The output
poses of the ligands generated were analyzed based on -CDOCKER
energies.
Pharmaceutical University, Nanjing 210009, People’s
Republic of China
Xiao-Qi Zhang − Institute of Traditional Chinese Medicine
and Natural Products, Jinan University, Guangzhou 510632,
People’s Republic of China; orcid.org/0000-0002-4436-
0273
Wen-Cai Ye − Institute of Traditional Chinese Medicine and
Natural Products, Jinan University, Guangzhou 510632,
People’s Republic of China; orcid.org/0000-0002-2810-
1001
Statistics. Experimental values were expressed as means
SD.
Data were analyzed using GraphPad Prism 8.0 software. The level of p
< 0.05 was used to determine the statistical significance.
Complete contact information is available at:
https://pubs.acs.org/10.1021/acs.jnatprod.1c00262
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge at
https://pubs.acs.org/doi/10.1021/acs.jnatprod.1c00262.
■
sı
Author Contributions
#X.-L.H. and Q.-W.H. contributed equally and are joint first
authors.
Supplementary methods and figures, chemical names,
synthetic yields, spectroscopic data, as well as NMR and
HRESIMS spectra of all synthetic compounds (PDF)
Notes
The authors declare no competing financial interest.
ACKNOWLEDGMENTS
AUTHOR INFORMATION
Corresponding Author
Hao Wang − State Key Laboratory of Natural Medicines,
School of Traditional Chinese Pharmacy, China
Pharmaceutical University, Nanjing 210009, People’s
Republic of China; orcid.org/0000-0003-3994-9806;
Email: wanghao@cpu.edu.cn
■
■
This work was supported financially by the National Natural
Science Foundation of China (nos. 81973206, U1801287, and
82073804), the China Postdoctoral Science Foundation (nos.
2019M662006 and 2019TQ0357), the Major National Science
and Technology Projects of the Chinese Thirteen Five-Year
Plan (no. 2017ZX09309024), the Local Innovative and
Research Teams Project of Guangdong Pearl River Talents
Program (No. 2017BT01Y036), the National Key R&D
Program of China (No. 2017YFC1703802), and the Jiangsu
Province Graduate Student Training Innovation Project (no.
KYLX16_1208).
Authors
Xiao-Long Hu − State Key Laboratory of Natural Medicines,
School of Traditional Chinese Pharmacy, China
Pharmaceutical University, Nanjing 210009, People’s
Republic of China
Qi-Wei He − State Key Laboratory of Natural Medicines,
School of Traditional Chinese Pharmacy, China
Pharmaceutical University, Nanjing 210009, People’s
Republic of China
Huan Long − State Key Laboratory of Natural Medicines,
School of Traditional Chinese Pharmacy, China
Pharmaceutical University, Nanjing 210009, People’s
Republic of China
Li-Xin Zhang − State Key Laboratory of Natural Medicines,
School of Traditional Chinese Pharmacy, China
Pharmaceutical University, Nanjing 210009, People’s
Republic of China
Rong Wang − State Key Laboratory of Natural Medicines,
School of Traditional Chinese Pharmacy, China
Pharmaceutical University, Nanjing 210009, People’s
Republic of China
Bao-Lin Wang − State Key Laboratory of Natural Medicines,
School of Traditional Chinese Pharmacy, China
Pharmaceutical University, Nanjing 210009, People’s
Republic of China
Jia-Hao Feng − State Key Laboratory of Natural Medicines,
School of Traditional Chinese Pharmacy, China
Pharmaceutical University, Nanjing 210009, People’s
Republic of China
Quan Wang − State Key Laboratory of Natural Medicines,
School of Traditional Chinese Pharmacy, China
Pharmaceutical University, Nanjing 210009, People’s
Republic of China
REFERENCES
■
(1) Salminen, A.; Lehtonen, M.; Paimela, T.; Kaarniranta, K.
Biochem. Biophys. Res. Commun. 2010, 394, 439−442.
(2) Yadav, P.; Jaswal, V.; Sharma, A.; Kashyap, D.; Tuli, H. S.; Garg,
V. K.; Das, S. K.; Srinivas, R. Pharm. Pat. Anal. 2018, 7, 155−167.
(3) Zhang, J.; Li, C. Y.; Xu, M. J.; Wu, T.; Chu, J. H.; Liu, S. J.; Ju,
W. Z. J. Ethnopharmacol. 2012, 144, 195−200.
(4) Sanna, V.; Chamcheu, J. C.; Pala, N.; Mukhtar, H.; Sechi, M.;
Siddiqui, I. A. Int. J. Nanomed. 2015, 10, 6835−6846.
(5) Figueiredo, S. A. C.; Salvador, J. A. R.; Cortes, R.; Cascante, M.
Eur. J. Med. Chem. 2017, 138, 422−437.
(6) Jiang, F.; Wang, H. J.; Bao, Q. C.; Wang, L.; Jin, Y. H.; Zhang,
Q.; Jiang, D.; You, Q. D.; Xu, X. L. Bioorg. Med. Chem. 2016, 24,
5431−5439.
(7) Xu, M.; Li, N.; Zhao, Z.; Shi, Z.; Sun, J.; Chen, L. Eur. J. Med.
Chem. 2019, 174, 265−276.
(8) Li, N.; Xu, M.; Wang, B.; Shi, Z.; Zhao, Z.; Tang, Y.; Wang, X.;
Sun, J.; Chen, L. J. Med. Chem. 2019, 62, 10798−10815.
(9) Ferrarini, M.; Heltai, S.; Zocchi, M. R.; Rugarli, C. Int. J. Cancer
1992, 51, 613−619.
(10) Kamal, A.; Boehm, M. F.; Burrows, F. J. Trends Mol. Med. 2004,
10, 283−290.
(11) Zhang, T.; Hamza, A.; Cao, X.; Wang, B.; Yu, S.; Zhan, C. G.;
Sun, D. Mol. Cancer Ther. 2008, 7, 162−170.
(12) He, Q. W.; Feng, J. H.; Hu, X. L.; Long, H.; Huang, X. F.; Jiang,
Z. Z.; Zhang, X. Q.; Ye, W. C.; Wang, H. J. Nat. Prod. 2020, 83,
2578−2586.
(13) Zhang, H. J.; Zhang, G. R.; Piao, H. R.; Quan, Z. S. J. Enzyme
Inhib. Med. Chem. 2018, 33, 190−198.
(14) Jafari, R.; Almqvist, H.; Axelsson, H.; Ignatushchenko, M.;
Lundback, T.; Nordlund, P.; Molina, D. M. Nat. Protoc. 2014, 9,
2100−2122.
Ji-Qin Hou − State Key Laboratory of Natural Medicines,
School of Traditional Chinese Pharmacy, China
1965
https://doi.org/10.1021/acs.jnatprod.1c00262
J. Nat. Prod. 2021, 84, 1954−1966

Journal of Natural Products
pubs.acs.org/jnp
Article
(15) Savitskaya, M. A.; Onishchenko, G. E. Biochemistry 2015, 80,
1393−1405.
(16) Wang, S.; Dong, Y. Y.; Liang, X. G. Biosens. Bioelectron. 2018,
109, 1−7.
(17) Xu, W. P.; Neckers, L. Clin. Cancer Res. 2007, 13, 1625−1629.
(18) Goel, S.; DeCristo, M. J.; McAllister, S. S.; Zhao, J. J. Trends
Cell Biol. 2018, 28, 911−925.
(19) Xu, X. L.; Wu, Y.; Hu, M. Y.; Li, X.; Bao, Q. C.; Bian, J. L.; You,
Q. D.; Zhang, X. J. Sci. Rep. 2016, 6, 35771.
(20) Jiang, F.; Wang, H. J.; Bao, Q. C.; Wang, L.; Jin, Y. H.; Zhang,
Q.; Jiang, D.; You, Q. D.; Xu, X. L. Bioorg. Med. Chem. 2016, 24,
5431−5439.
(21) Hu, X. L.; Feng, J. H.; Pham, T. A.; Ma, H. Y.; Ma, M. X.; Rui,
S.; Wei, S.; Fei, X.; Zhang, X. Q.; Ye, W. C.; Hao, W. Phytomedicine
2018, 50, 88−98.
(22) Gao, C. L.; Hou, G. G.; Liu, J.; Ru, T.; Xu, Y. Z.; Zhao, S. Y.;
Ye, H.; Zhang, L. Y.; Chen, K. X.; Guo, Y. W.; Pang, T.; Li, X. W.
Angew. Chem., Int. Ed. 2020, 59, 2429−2439.
(23) Huang, J. J.; Xie, Y.; Li, H.; Zhang, X. X.; Huang, Q.; Zhu, Y.;
Gu, P.; Jiang, W. M. Chin. J. Nat. Medicines 2019, 17, 746−755.
1966
https://doi.org/10.1021/acs.jnatprod.1c00262
J. Nat. Prod. 2021, 84, 1954−1966