Hydroxamic Acid Derivatives of β-Carboline/Hydroxycinnamic Acid Hybrids Inducing Apoptosis and Autophagy through the PI3K/Akt/mTOR Pathways

Article abstract of DOI:10.1021/acs.jnatprod.8b00843

Naturally occurring β-carbolines are known to have antitumor activities but with limited effectiveness. In order to improve their efficacy, a series of new hydroxamic-acid-containing β-carbolines connected via a hydroxycinnamic acid moitey (12a-f) were developed to incorporate histone deacetylase (HDAC) inhibition for possible synergistic effects. When evaluated in in vitro assays, most of the analogues showed significant antitumor activities against four human cancer cells. In particular, 12b showed the highest cytotoxic potency of the series, including drug-resistant Bel7402 cells, but had minimal effect on normal hepatic LO2 cells. These compounds also showed excellent inhibitory effects against HDAC1/6, which appear to contribute greatly to their antiproliferative properties. Compound 12b enhanced the acetylation levels of histone H3 and α-tubulin and induced greater cancer cell apoptosis than the FDA-approved HDAC inhibitor SAHA by regulating expression of apoptotic proteins Bax, Bcl-2, and caspase 3. Importantly, 12b also induced a significant amount of autophagic flux activity in Bel7402 cells by increasing the expression of Beclin-1 and LC3-II proteins and decreasing that of LC3-I and p62. Finally, 12b significantly inhibited PI3K/Akt/mTOR signaling, an important cell-growth-promoting pathway aberrantly activated in many cancers. Together, the results suggest that these hydroxamic-acid-containing β-carboline derivatives may be new leads for the discovery of agents for the treatment of human carcinoma cancers.

Full text of DOI:10.1021/acs.jnatprod.8b00843

Article
pubs.acs.org/jnp
Cite This: J. Nat. Prod. XXXX, XXX, XXX−XXX
Hydroxamic Acid Derivatives of β‑Carboline/Hydroxycinnamic Acid
Hybrids Inducing Apoptosis and Autophagy through the PI3K/Akt/
mTOR Pathways
Yong Ling,*,†,‡ Yangyang Li,†,‡ Rui Zhu,† Jianqiang Qian,† Ji Liu,†,‡ Weijie Gao,† Chi Meng,†
Jiefei Miao,†,‡ Biao Xiong,† Xiaodong Qiu,† Changchun Ling,†,‡ Hong Dai,*,§ and Yanan Zhang*,†
†School of Pharmacy and Jiangsu Province Key Laboratory for Inflammation and Molecular Drug Target, Nantong University,
Nantong 226001, People’s Republic of China
‡The Affiliated Hospital of Nantong University, Nantong University, Nantong 226001, People’s Republic of China
§College of Chemistry and Chemical Engineering, Nantong University, Nantong 226019, People’s Republic of China
S
* Supporting Information
ABSTRACT: Naturally occurring β-carbolines are known to have antitumor activities but with limited effectiveness. In order to
improve their efficacy, a series of new hydroxamic-acid-containing β-carbolines connected via a hydroxycinnamic acid moitey
(12a−f) were developed to incorporate histone deacetylase (HDAC) inhibition for possible synergistic effects. When evaluated
in in vitro assays, most of the analogues showed significant antitumor activities against four human cancer cells. In particular,
12b showed the highest cytotoxic potency of the series, including drug-resistant Bel7402 cells, but had minimal effect on normal
hepatic LO2 cells. These compounds also showed excellent inhibitory effects against HDAC1/6, which appear to contribute
greatly to their antiproliferative properties. Compound 12b enhanced the acetylation levels of histone H3 and α-tubulin and
induced greater cancer cell apoptosis than the FDA-approved HDAC inhibitor SAHA by regulating expression of apoptotic
proteins Bax, Bcl-2, and caspase 3. Importantly, 12b also induced a significant amount of autophagic flux activity in Bel7402
cells by increasing the expression of Beclin-1 and LC3-II proteins and decreasing that of LC3-I and p62. Finally, 12b
significantly inhibited PI3K/Akt/mTOR signaling, an important cell-growth-promoting pathway aberrantly activated in many
cancers. Together, the results suggest that these hydroxamic-acid-containing β-carboline derivatives may be new leads for the
discovery of agents for the treatment of human carcinoma cancers.
communicable diseases, after cardiovascular disease. The
number of cancer deaths accounts for about 13% of the
world’s total deaths, and more than 8.2 million people died of
cancer in 2014.6 Chemotherapy remains the principal
treatment for most human cancers; however, many of the
current chemotherapeutic agents suffer from limitations
including systemic toxicity, low selectivity, and drug resist-
ance,7−9 making the discovery of new chemodrugs an urgent
need.
atural products have historically been a main source of
Ndrugs for the treatment of various diseases including
cancers. Approximately 25% of all current drugs are derived
from plants, and many others are synthetic analogues built
upon prototype compounds isolated from plants.1,2 Many
medicinal plant-based medicines, such as the camptothecin
derivatives and taxanes (e.g., Taxol), are still an essential part
of the armament for treating cancers, and natural products
continue to provide leads for drug candidates that have entered
clinical trials as anticancer agents.3,4
Despite these advances, malignant tumors remain a serious
threat to human health.5 According to the WHO, cancer is the
second most frequent cause of death among the non-
Received: October 11, 2018
© XXXX American Chemical Society and
American Society of Pharmacognosy
A
DOI: 10.1021/acs.jnatprod.8b00843
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Naturally occurring β-carbolines, including harmine, harma-
line, harmalol, and harmane, are indole alkaloids with a planar
tricyclic 9H-pyrido[3,4-b]indole skeleton (Figure 1). Originally
for the treatment of multiple cancers, particularly cutaneous T
cell lymphoma (CTCL),24 but are generally less effective
toward solid tumors. Given the antitumor effects of β-
carbolines on DNA damage and drug resistance, incorporation
of an HDAC inhibitory hydroxamic acid fragment into the
natural β-carbolines may lead to hybrid molecules with
increased anticancer potency/efficacy or even reversal of
drug resistance to a broader range of cancer, even solid tumors.
Finally, the hydroxamic acid functionality from HDAC
inhibitors could readily be connected to β-carbolines using a
hydroxycinnamic acid (HCA) structural moiety. HCAs
including ferulic acid (FA) and 4-hydroxycinnamic acid
(Figure 2), widely occur in plants and show antiproliferative
activity against several types of cancer cells.25−27 HCAs and
their derivatives have also been shown to block the
transcription factor (NF-κB) pathway and trigger an apoptotic
cascade.28,29 In addition, N-hydroxycinnamamide, the hydroxa-
mic acid analogue of HCA, is present in several FDA-approved
HDAC inhibitors such as belinostat (PXD101) and panobino-
stat (LBH589) (Figure 2),30,31 thus making HCA an ideal
connector for the hybrids. In addition to linking β-carbolines
and the hydroxamic acid functionality, the HCA moiety may
potentially contribute to the anticancer effects of the hybrids.
Herein, the synthesis and biological evaluation of the new β-
carboline/HCA hybrids (12a−f, Figure 3), including the
Figure 1. Structures of natural β-carbolines harmine, harmaline,
harmalol, and harmane.
isolated from the plant Peganum harmala (Syrian rue, family
Zygophyllaceae), β-carbolines are also known as harmala
alkaloids, occurring in Peganum harmala in concentrations of
2−7% or even higher.10 The crude extract of Peganum harmala
has been traditionally used as an herbal medicine in Central
Asia, the Middle East, and North Africa because of its
antibacterial effect and also in clinical treatment of cancer due
to its inhibitory effects of cancer cells.11,12 Indeed, β-carbolines
have been indicated in a range of diverse biological processes,
including antitumor, antiviral, antimicrobial, and antiparasitic
activities.13−15 β-Carbolines have attracted attention for the
treatment of cancer due to their intercalation into DNA,
inhibition of CDK and topoisomerase, and antiangiogenesis
activity.16−18 In particular, β-carbolines such as harmine, were
found to reduce resistance to several common anticancer drugs
such as mitoxantrone and camptothecin.18 However, these β-
carbolines show only modest efficacy as anticancer drugs and
have often been used in combination with other anticancer
agents to improve effectiveness.19,20
Multitarget drugs comprising hybrids that combine struc-
tural components of two or more active drugs have been a
promising strategy to generate potential leads that target
multiple signaling pathways, therefore leading to improved
biological activities and therapeutic efficacies.21,22 This
approach is consistent with the traditional use of medicinal
plants in cancer treatment, which, unlike pharmacological
drugs, commonly contain a mixture of components that may
act simultaneously on multiple protein targets. To this end, we
envisioned the introduction of a hydroxamic acid functionality
to the β-carboline, the former being an active fragment in
many histone deacetylase (HDAC) inhibitors including
suberanilohydroxamic acid (vorinostat, SAHA), an FDA
approved drug (Figure 2). HDACs are a class of promising
therapeutic targets for cancer intervention due to their
involvement in gene transcription, cell cycle progression, and
cell apoptosis and their overexpression in a variety of human
tumors.23 A variety of HDAC inhibitors have been approved
Figure 3. Hybrids based on β-carboline/HCA with hydroxamic acid
functionality.
investigation of their potency and antitumor mechanisms such
as tumor cell apoptosis induction and autophagy in several
cancer cell lines, are reported. In view of the above discussion,
we hypothesized that these hybrids, formed via conjugating the
natural β-carboline with a hydroxamic acid using HCA as a
linker, could display synergistic effects and improve anticancer
activities.
RESULTS AND DISCUSSION
The tricyclic ring system of β-carboline alkaloids is commonly
constructed by a Pictet−Splengler reaction between trypto-
■
Figure 2. Hydroxycinnamic acids (HCA) and representative hydroxamic-acid-containing HDAC inhibitors.
B
DOI: 10.1021/acs.jnatprod.8b00843
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
a
Scheme 1
a
Reagents and conditions: (a) thionyl chloride, MeOH, 0 °C, 1 h, and then 65 °C, 4 h; (b) ω-dibromoalkane, K2CO3, CH3CN, reflux, 4 h; (c) 4-
methoxybenzaldehyde, HOAc, reflux, 4 h; (d) thionyl chloride, MeOH, 0 °C, 1 h, and then 65 °C, 6 h; (e) KMnO4, DMF, rt, 5 h; (f) hydrazine
monohydrate, MeOH, 50 °C, 6 h; (g) NaNO2, HCl; (h) HOAc, H2O, 50 °C, 6 h; (i) K2CO3, CH3CN, reflux, 12 h; (j) NH2OK, MeOH, rt, 8−12
h.
Table 1. IC50 Values of 12a−f against Four Human HCC Cells and HDAC1
a
in vitro antiproliferative activity (IC50 , μM)
compd
SMMC-7721
HepG2
Bel7402
Huh7
HDAC1 (IC50, nM)
SAHA
harmine
12a
12b
12c
12d
12e
12f
5.2 0.5
47.6 5.1
4.0 0.5
2.9 0.3
7.1 0.6
6.3 0.7
4.1 0.4
>12.5
5.9 0.7
53.6 5.5
4.8 0.6
1.4 0.3
5.3 0.6
5.8 0.6
3.4 0.4
9.5 1.1
4.7 0.6
ND
5.1 0.6
ND
2.8 0.3
1.1 0.2
6.2 0.2
ND
142 18
>1000
b
1.8 0.2
1.0 0.1
4.6 0.5
5.2 0.6
4.0 0.5
8.9 1.0
32
27
61
75
57
3
3
7
8
7
ND
ND
125 13
a
b
Related data appeared as the means SD of three separate assays. ND: not detected.
phan and aldehydes. This strategy was used to prepare target
compounds 12a−f as outlined in Scheme 1. First, the
cinnamates 3a−f were prepared by esterification of hydrox-
ycinnamic acid 1a or 1b with SOCl2 in CH3OH to afford
methyl esters 2a,b. Treatment with ω-dibromoalkanes in the
presence of K2CO3 afforded the appropriate ethers 3a−f. On
the other hand, L-tryptophan (4) was converted to
tetrahydrocarboline 5 in a Pictet−Spengler reaction with 4-
methoxybenzaldehyde. Compound 5 was esterified by treat-
ment with SOCl2 in CH3OH to give methyl ester 6, which was
subsequently oxidized with KMnO4 in dimethylformamide
(DMF) to afford compound 7. Compound 7 was reacted with
hydrazine hydrate to afford hydrazide 8. Hydrazide 8 was
transformed to acyl azide 9 using NaNO2. Curtis rearrange-
ment of acyl azide 9 was carried out in aqueous HOAc to give
compound 10. Reaction between 10 and bromides 3a−f in the
presence of K2CO3 in CH3CN afforded the key intermediates
11a−f. Finally, treatment of 11a−f with NH2OK afforded the
target compounds 12a−f. These compounds were purified by
silica gel column chromatography, and their structures were
1
validated by MS, H NMR, 13C NMR, and HRMS data. The
compounds had a purity of >95% as determined by HPLC.
In order to obtain their cancer cell growth inhibitory activity,
the target compounds 12a−f were first tested against four
human hepatocellular carcinoma (HCC) cells (Bel7402,
SMMC-7721, Huh7, and HepG2) in MTT assays. These
cells were chosen to investigate how these hybrids would
inhibit the growth of solid tumors, against which HDAC
inhibitors are generally not effective. Since harmine has been
shown to reduce resistance to some anticancer drugs, drug-
sensitive Bel7402 cells, which are innately resistant to
chemodrugs including doxorubicin and vincristine,32 were
C
DOI: 10.1021/acs.jnatprod.8b00843
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
also included. Harmine and the FDA-approved HDAC
inhibitor for CTCL, SAHA, were tested in parallel. As shown
in Table 1, compounds 12a−f were all good antiproliferative
agents, displaying IC50 values in the low micromolar range in
all four human cancer cell lines. These IC50 values are
significantly lower than that of harmine, and most of them
were similar to or even lower than SAHA. Particularly, 12b had
IC50 values of 1.0 and 1.1 μM in Bel7402 and Huh7 cells,
respectively, which were 5-fold lower than those of SAHA
(IC50 = 4.7 and 5.1 μM).
activities contribute greatly to the antiproliferative properties
of this series of hybrids.
Since other members of the HDAC family have been
indicated in cancers, compounds 12a and 12b, two of the most
potent compounds of the series inhibiting HDAC1, were
tested for their potencies against HDAC3, -6, and -8 by
determining the fluorescent-based activity using recombinant
human enzymes. SAHA was again tested in parallel. As
summarized in Table 2, 12a and 12b both showed good
It appears that compounds 12a−c based on ferulic acid (R =
OMe) were more potent in antiproliferation than the
corresponding 4-hydroxycinnamic-acid-based 12d−f (R =
H). As for the methylene groups connecting the β-carboline
and HCA, compounds with propyl (n = 3) groups exhibited
greater antitumor activities than for ethyl (n = 2) and butyl (n
= 4) analogues.
Compound 12b was the most active compound of the series
and was further evaluated for its selectivity profile against
normal cells by determining its growth inhibitory effects on
drug-sensitive HCC Bel7402 cells, drug-resistant HCC
Bel7402/5-FU cells, and normal liver LO2 cells. Treatment
with an increasing dose of 12b displayed little effect on the
survival of nontumor LO2 cells; however, the same treatment
induced substantial inhibition of both Bel7402 and Bel7402/5-
FU cell proliferation (Figure 4). These findings suggest that
12b is selective in the antiproliferation of tumor cells over
normal cells.
Table 2. IC50 Values of Compounds 12a,b against HDAC1,
HDAC3, HDAC6, and HDAC8
a
IC50 (nM)
compound
HDAC1
HDAC3
HDAC6
HDAC8
SAHA
12a
12b
142 18
153 17
125 15
148 16
65
17
13
7
3
2
425 51
>1000
>1000
32
27
3
3
a
These data are from three independent experiments and are
expressed as the mean SD.
inhibitory potencies against HDAC6 with IC50 values of 16
and 13 nM, respectively. These values are about 2-fold lower
than HDAC1, and 7- to 11-fold lower than HDAC3. Neither
12a nor 12b showed much activity against HDAC8 (IC50
>
1000 nM). Importantly, the IC50 values of 12b were about 5-
fold lower than those of SAHA on both HDAC1 and HDAC6.
HDACs deacylate histones and suppress gene expression,
and therefore inhibiton of HDACs at the cellular level would
increase acetylation of histones. Expression levels of acetylated
histone H3 and α-tubulin, biomarkers of HDAC inhibition,
were determined upon treatment with 12b in Western blot
assays (Figure 5). Bel7402 cells were thus incubated with
either the vehicle, SAHA (5.0 μM), or 12b (1.0 and 3.0 μM)
for 72 h. Compound 12b significantly enhanced histone H3
Figure 4. Inhibitory effects of 12b on the proliferation of Bel7402,
Bel7402/5-FU, and LO2 cells. Cells were incubated with different
concentrations of 12b for 72 h. Cell inhibitory % was determined
using the MTT assay. These data are from three independent
experiments and are presented as the mean SD.
All targets were tested for their inhibitory effects on HDAC1
protein in the enzymatic fluorimetric activity assay. HDAC1 is
a member of the HDAC family important for cell proliferation
and tissue development and is routinely targeted for cancer
treatment. While harmine showed little activity at HDAC1
(IC50 > 1000 nM), compounds 12a−f all showed potencies
similar to or greater than SAHA (Table 1). This confirms that
the presence of the hydroxamic acid induced HDAC activities.
Interestingly, the relative HDAC1 potencies matched well with
their cell-based antiproliferation results. The lower potencies of
these compounds in cell antiproliferation, which were in the
μM range, compared to nM potencies in HDAC inhibition, are
likely the result of their interactions with other proteins present
in the cells, rendering lowered effective concentrations for
HDACs. These results suggest that the HDAC inhibitory
Figure 5. Immunoblot analysis of Ac-histone H3 and Ac-α-tubulin in
vitro. (A) Western blot analysis of Bel7402 cells treated with 12b and
SAHA. (B) Quantitative analysis. Related data appear as means SD
of three separate assays. *P < 0.01 vs control.
D
DOI: 10.1021/acs.jnatprod.8b00843
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Figure 6. Compound 12b induces Bel7402 cell apoptosis in vitro. (A) Flow cytometry analysis of Bel7402 cells treated with 12b, SAHA, or
harmine, respectively, and stained with FITC-annexin V/PI. (B) Quantitative analysis of percentages of apoptotic cells. Related data appear as
means SD of three separate assays. *P < 0.01 vs control.
Figure 7. (A) Expression levels of Bcl-2, Bax, cleaved-caspase 3, cleaved-PARP, and β-actin as determined by Western blot analysis using Bel7402
cells treated with vehicle, 12b, harmine, or SAHA. (B) Relative levels of each protein (Bcl-2, Bax, cleaved-caspase 3, and cleaved-PARP) compared
to β-actin. Related data appear as the means SD of three separate assays. *P < 0.01 vs control.
acetylation and also increased the Ac-α-tubulin expression level
in a dose-dependent fashion, suggesting HDAC inhibiton. The
Ac-histone H3 and Ac-α-tubulin levels in the 12b-treated
groups were similar (1.0 μM) or even higher (3.0 μM) than
the 5.0 μM SAHA-treated group, agreeing with their HDAC
isoform enzymatic activities.
evaluate 12b for its apoptosis induction effect, the expression
levels of apoptotic proteins Bax and Bcl-2 and the cleavage
states of caspase-3 and poly ADP ribose polymerase (PARP, an
apoptosis marker)37 upon treatment with 12b (Figure 7) were
evaluated. Subconfluent Bel7402 HCC cells were treated with
vehicle or 12b for 72 h, and the cells were lysed and subjected
to analysis by Western blot. As shown in Figure 7A, treatment
with 12b markedly increased the proapoptotic Bax levels but
decreased the antiapoptotic Bcl-2 levels in a dose-dependent
fashion. Additionally, compound 12b led to cleavage of both
caspase-3 and PARP, both more significant than the control
group and the SAHA-treated group. Accordingly, these
immunohistochemical analyses confirmed that treatment with
12b induced tumor cell apoptosis by regulating apoptotic-
related protein levels in Bel7402 cells.
Autophagy, the natural, regulated process that the cell uses
to disassemble unnecessary or dysfunctional components, is
considered a promising and novel strategy to enhance
antitumor efficacy of chemotherapy drugs.38,39 To investigate
whether 12b induced autophagy in Bel7402 cells, acridine
orange (AO) staining was used to detect autophagic formation
using fluorescence microscopy. As shown in Figure 8, in
control cells green fluorescence was primarily emitted with
only weak red fluorescence, indicating a low autophagic
induction. On the other hand, Bel7402 cells treated with 12b
Apoptosis is a process critically involved in tumor formation
and treatment responses.33−35 Since it showed significant HCC
cell antiproliferative activities and also selectively inhibited
HDAC1/6, 12b was assessed for its ability to enhance cancer
cell apoptosis. Thus, Bel7402 cells were treated with either
SAHA, harmine, or 12b for 72 h, and the apoptotic rates of
cells were analyzed using FITC-annexin V/PI staining and flow
cytometry. As shown in Figure 6, cells exposed to increasing
concentrations of 12b displayed higher percentages of annexin
V + apoptotic Bel7402 cells (49.6% for 1.0 μM and 73.8% for
5.0 μM). These results were higher than the SAHA (45.4% for
5.0 μM) and harmine (36.1% for 20 μM) groups at the same
or higher concentrations. Therefore, 12b induced Bel7402 cell
apoptosis with greater potency than both SAHA and harmine.
Caspase-3 is a member of the cysteine−aspartic acid
protease (caspase) family and is closely associated with
apoptosis. In most cells caspase 3 is normally expressed in
an inactive pro-enzyme form, but upon activation, it can often
activate other procaspases, triggering a protease cascade.36 To
E
DOI: 10.1021/acs.jnatprod.8b00843
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
lators.41 To further study the underlying molecular mecha-
nisms of 12b, the effects of 12b on regulating PI3K/Akt/
mTOR signaling in Bel7402 cells, using SAHA and harmine as
controls, were investigated. Thus, Bel7402 cells were treated
with harmine, SAHA, or 12b at 1.0 and 2.5 μM, and the
expression and activation of Akt and mTOR were analyzed by
Western blotting. Interestingly, treatment with 12b signifi-
cantly inhibited the phosphorylation of both Akt and mTOR in
a dose-dependent manner, whereas it did not change the
expression levels of these two proteins in these cancer cells
(Figure 10). Similarly, treatment with SAHA and harmine also
effectively reduced the levels of Akt and mTOR phosphor-
ylation, although to a lesser extent compared to 12b. These
results suggest that treatment with 12b could significantly
inhibit PI3K/Akt/mTOR signaling, thus generating high-
lighted anticancer effects.
In summary, we report a new series of hydroxamic-acid-
based β-carboline/hydroxycinnamic acid hybrids, 12a−f,
which incorporate the key structural components of harmine,
classical HDACIs, and HCA. While HDAC inhibitors are
generally not effective against solid tumors and β-carboline
possesses weak anticancer activities, a combination of these
components may lead to synergistic effects and thus improved
efficacy or even drug resistance reversal. These hybrids were
evaluated for their biological activities and anticancer
mechanisms in a battery of in vitro assays. Most of the
compounds displayed significant antiproliferative activity with
IC50 values in the low micromolar range. In particular, 12b
selectively inhibited tumor cell proliferation of drug-sensitive
Bel7402 and drug-resistant Bel7402/5-FU cells, but not
normal LO2 cells. These hybrids appeared to retain HDAC
inhibitory effects, and both 12a and 12b exerted selective
HDAC1/6 inhibition. Treatment with 12b resulted in
enhancement of acetylated histones and acetylated α-tubulin
protein levels and promoted cell apoptosis by enhancing the
expression of cleaved caspase-3 and PARP in Bel7402 cells. In
addition, 12b also induced a significant amount of autophagic
flux activity in Bel7402 cells by the upregulation of LC3-II and
Beclin-1 proteins and downregulation of p62 and LC3-I.
Finally, 12b could significantly inhibit PI3K/Akt/mTOR
signaling, thus generating highlighted anticancer effects.
Given their antitumor activities, particularly against drug-
sensitive and -resistant cancer cells, these β-carboline/
hydroxamic acid hybrids may warrant further evaluation as
potential candidates for the treatment of human cancer.
Figure 8. Detection of autophagic vacuoles in Bel7402 cells after
various treatments for 24 h: (a) control cells; (b) harmine at 20 μM;
(c) SAHA at 5.0 μM; (d) 1.0 μM 12b-treated cells; and (e) 5.0 μM
12b-treated cells. The level of acridine orange staining was
determined by fluorescence microscopy. Representative images were
captured from three independent experiments under 400× magnifi-
cation.
displayed an increase in red fluorescence with the increase of
drug concentration, while the green fluorescence intensity
decreased. These results suggest a significant amount of
autophagic flux activity in 12b-treated groups in Bel7402 cells,
more significant than the harmine- and SAHA-treated groups.
Next, the ability and mechanism of 12b in induction of
autophagy were further investigated by determining the levels
of several important autophagy-associated proteins. The
transformation of LC3-I into LC3-II is a key step in autophagy,
and the number of LC3-II puncta correlates with the number
of autophagosomes. Thus, the expression of LC3-I, LC3-II,
p62, and Beclin-1, proteins all related to autophagy, in
response to 12b treatment was examined (Figure 9). Western
blot assay showed a substantial increase in LC3-II and Beclin-1
protein expression levels, whereas the expression of p62 and
LC3-I was reduced upon treatment with 12b of different
concentrations. The expression of LC3-II in 12b-treated
groups was higher than that of the SAHA-treated group
(Figure 9B).
The PI3K/Akt/mTOR pathway is an essential signaling
pathway in the regulation of the cell cycle, and its aberrant
activation promotes tumor cell survival and proliferation in
many cancers.40 Once PI3K activates the serine/threonine
kinase Akt, Akt will then phosphorylate and activate the
serine/threonine kinase mTOR through a series of regu-
Figure 9. (A) Expression levels of autophagy-related proteins LC3, p62, and beclin-1 as determined by Western blot analysis in Bel7402 cells
treated with vehicle, 12b, harmine, or SAHA. (B) Quantitative analysis of LC3, p62, and beclin-1 levels relative to control β-actin. Related data
appeared as the means SD of three separate assays. *P < 0.01 vs respective control.
F
DOI: 10.1021/acs.jnatprod.8b00843
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Figure 10. Effects of 12b on the PI3K/Akt/mTOR signaling in Bel7402 cells. (A) Levels of Akt and mTOR expression and phosphorylation
relative to β-actin by Western blot analysis. (B) Quantitative analysis. Related data appear as the means SD of three separate assays. *P < 0.01 vs
the vehicle-treated control.
122.0, 121.3, 118.9, 117.5, 113.5, 112.5, 69.2, 57.0, 56.3, 42.9; HRMS
EXPERIMENTAL SECTION
■
(ESI) m/z calcd for C30H29N4O5 [M + H]+, 525.2138; found,
1
General Experimental Procedures. H and 13C NMR spectra
525.2153.
were collected on a Bruker AV 400 M spectrometer in DMSO-d6 or
Compound 12b. Refer to 12a; compound 12b was synthesized
CDCl3, and tetramethylsilane was used as the internal standard. Mass
starting from 10, 3b, and NH2OK in a yield of 46%. MS (ESI) m/z =
spectra was recorded on a Mariner mass spectrometer (ESI). HRMS
539 [M + H]+; 1H NMR (DMSO-d6, 400 MHz) δ 10.62 (s, 1H, NH),
data were acquired with an Agilent Technologies LC/MSD TOF.
8.96 (s, 1H, NH), 8.02 (m, 4H, Ar-H), 7.45 (m, 4H, NH, Ar-H, CH),
HCA (1), L-tryptophan (4), and 4-methoxybenzaldehyde were
7.13 (m, 6H, Ar-H), 6.34 (d, 1H, J = 16.0 Hz, CH), 4.16 (m, 2H,
purchased from Aladdin (Shanghai, China). Compounds 2 and 6−
CH2), 3.85 (m, 6H, OCH3), 3.50 (m, 2H, CH2), 2.11 (m, 2H, CH2);
10 were prepared according to literature procedures.42
13C NMR (DMSO-d6, 100 MHz) δ 167.1, 151.8, 149.7, 149.4, 145.2,
Preparation of 3a−f. Compound 3a. To a solution of 1,2-
143.1, 141.6, 138.8, 134.6, 132.1, 130.5, 130.0, 129.5, 128.5, 127.7,
dibromoethane (2.78 g, 15 mmol) were added compound 2a (2.08 g,
10 mmol) and K2CO3 (2.75 g, 20 mmol). The mixture was stirred at
127.3, 126.0, 123.9, 121.6, 121.1, 119.0, 118.3, 114.2, 112.4, 68.0,
56.3, 55.8, 39.6, 30.4; HRMS (ESI) m/z calcd for C31H31N4O5 [M +
70 °C for 4 h. The mixture was concentrated, and the residue was
H]+, 539.2294; found, 539.2307.
collected through extraction and concentration and further purified
through column chromatography (petroleum ether/EtOAc = 15:1) to
Compound 12c. Refer to 12a; compound 12c was obtained
starting from 10, 3c, and NH2OK in a yield of 49%; MS (ESI) m/z =
afford a colorless oil. Yield: 73.1%. MS (ESI) m/z = 315 [M + H]+.
553 [M + H]+; 1H NMR (DMSO-d6, 400 MHz) δ 10.76 (s, 1H, NH),
Compound 3b. Refer to 3a; compound 3b was synthesized starting
8.00 (m, 3H, Ar-H), 7.43 (m, 5H, NH, Ar-H, CH), 7.09 (m, 6H, Ar-
from 2a and 1,3-dibromopropane to give a light yellow oil. Yield:
68.0%. MS (ESI) m/z = 329 [M + H]+.
H, NH), 6.55 (d, 1H, J = 16.0 Hz, CH), 5.98 (s, 1H, Ar-H), 4.07
(m, 2H, CH2), 3.85 (m, 6H, OCH3), 3.39 (m, 3H, OH, CH2), 1.76−
1.89 (m, 4H, 2 × CH2); 13C NMR (DMSO-d6, 100 MHz) δ 167.3,
151.8, 149.7, 149.4, 143.5, 141.9, 140.6, 138.8, 133.9, 133.1, 132.1,
130.5, 129.0, 128.5, 128.2, 127.8, 125.9, 123.3, 121.5, 121.1, 119.0,
117.7, 114.8, 112.3, 63.9, 56.3, 55.8, 38.7, 29.0, 23.0; HRMS (ESI) m/
z calcd for C32H33N4O5 [M + H]+, 553.2451; found, 553.2442.
Compound 12d. Refer to 12a; compound 12d was obtained
starting from 10, 3d, and NH2OK in a yield of 55%. MS (ESI) m/z =
495 [M + H]+; 1H NMR (DMSO-d6, 400 MHz) δ 10.63 (s, 1H, NH),
8.01 (d, 1H, J = 4 Hz, Ar-H), 7.92 (m, 2H, Ar-H), 7.68 (m, 2H, Ar-
H), 7.45 (m, 5H, NH, Ar-H, CH), 7.10 (m, 4H, Ar-H, NH), 6.60 (m,
3H, Ar-H, CH), 4.27 (m, 2H, CH2), 3.85 (m, 4H, OCH3, OH), 3.76
(m, 2H, CH2); 13C NMR (DMSO-d6, 100 MHz) δ 167.5, 152.8,
151.9, 145.5, 143.6, 143.3, 138.9, 134.1, 133.8, 133.3, 132.9, 132.0,
130.3, 130.0, 129.3, 128.6, 128.0, 121.5, 119.2, 114.8, 112.8, 69.1,
56.2, 43.9; HRMS (ESI) m/z calcd for C29H27N4O4 [M + H]+,
495.2032; found, 495.2023.
Compound 12e. Refer to 12a; compound 12e was obtained
starting from 10, 3e, and NH2OK in a yield of 56%. MS (ESI) m/z =
509 [M + H]+; 1H NMR (DMSO-d6, 400 MHz) δ 10.81 (s, 1H, NH),
8.07 (d, 1H, J = 8.0 Hz, Ar-H), 7.68 (m, 3H, Ar-H), 7.49 (m, 7H,
NH, Ar-H, CH), 7.14 (m, 3H, Ar-H, NH), 6.68 (m, 3H, Ar-H, CH),
4.19 (m, 2H, CH2), 3.85 (s, 3H, OCH3), 3.50 (m, 2H, CH2), 2.13
(m, 2H, CH2); 13C NMR (DMSO-d6, 100 MHz) δ 165.7, 152.6,
151.9, 146.3, 143.4, 142.1, 139.9, 138.9, 134.1, 132.8, 132.4, 131.5,
130.4, 129.8, 129.2, 128.6, 127.9, 122.6, 121.8, 120.7, 113.9, 113.3,
67.7, 55.9, 38.2, 30.7; HRMS (ESI) m/z calcd for C30H29N4O4 [M +
H]+, 509.2189; found, 509.2201.
Compound 3c. Refer to 3a; compound 3c was synthesized starting
from 2a and 1,4-dibromobutane to give a light yellow oil. Yield:
70.1%. MS (ESI) m/z = 343 [M + H]+.
Compound 3d. Refer to 3a; compound 3d was synthesized starting
from 2b and 1,2-dibromoethane to give a light yellow oil. Yield:
75.2%. MS (ESI) m/z = 285 [M + H]+.
Compound 3e. Refer to 3a; compound 3e was synthesized starting
from 2b and 1,3-dibromopropane to give a light yellow oil. Yield:
69.4%. MS (ESI) m/z = 299 [M + H]+.
Compound 3f. Refer to 3a; compound 3f was synthesized starting
from 2b and 1,4-dibromobutane to give a light yellow oil. Yield:
66.1%. MS (ESI) m/z = 313 [M + H]+.
Preparation of 12a−f. Compound 12a. Compound 10 (2.89 g,
10 mmol), K2CO3 (2.75 g, 20 mmol), and 3a (3.14 g, 10 mmol) were
added to CH3CN (40 mL), and the mixture was stirred at 55 °C for
12 h. The mixture was concentrated, and the residue was diluted with
H2O and extracted with EtOAc (2 × 50 mL). The combined EtOAc
solvent was dried over Na2SO4, filtered, and concentrated. The
residue was dissolved in a solution of NH2OK (2.8 g, 40 mmol) in
anhydrous MeOH (50 mL) and stirred for 8−12 h. Upon completion
of the reaction, the mixture was concentrated, and the residue was
collected through extraction and concentration and further purified
through quick column chromatography to produce a light yellow
solid. Yield: 52%. MS (ESI) m/z = 525 [M + H]+; 1H NMR (DMSO-
d6, 400 MHz) δ 11.75 (s, 1H, NH), 8.63 (s, 1H, NH), 8.39 (d, 1H, J
= 8.0 Hz, Ar-H), 7.96 (m, 4H, Ar-H), 7.52 (m, 7H, Ar-H, NH, CH),
6.96 (m, 2H, Ar-H), 6.53 (d, 1H, J = 16.0 Hz, Ar-H), 4.27 (m, 2H,
CH2), 3.88 (s, 6H, OCH3), 3.76 (m, 2H, CH2); 13C NMR (DMSO-
d6, 100 MHz) δ 167.4, 151.3, 149.9, 149.0, 146.0, 143.1, 142.1, 139.8,
134.8, 133.7, 131.3, 130.3, 128.6, 128.0, 127.5, 127.2, 126.0, 123.0,
Compound 12f. Refer to 12a; compound 12f was obtained starting
from 10, 3f, and NH2OK in a yield of 51%. MS (ESI) m/z = 523 [M
G
DOI: 10.1021/acs.jnatprod.8b00843
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
+ H]+; 1H NMR (DMSO-d6, 400 MHz) δ 10.76 (s, 1H, NH), 7.97−
8.04 (m, 3H, Ar-H), 7.35−7.53 (m, 6H, NH, Ar-H, CH), 7.01−7.13
(m, 5H, Ar-H, NH), 6.77 (m, 1H, Ar-H), 6.57 (d, 1H, J = 16.0 Hz, 
CH), 4.18 (m, 2H, CH2), 3.86 (s, 3H, OCH3), 3.50 (m, 2H, CH2),
1.79−1.90 (m, 4H, 2 × CH2); 13C NMR (DMSO-d6, 100 MHz) δ
165.8, 152.6, 151.8, 145.3, 143.1, 142.1, 140.3, 138.2, 134.5, 132.6,
131.6, 130.2, 129.2, 128.8, 128.2, 128.0, 127.9, 122.6, 121.8, 120.8,
116.3, 113.2, 64.1, 56.1, 38.1, 28.6, 22.9; HRMS (ESI) m/z calcd for
C31H31N4O4 [M + H]+, 523.2345; found, 523.2353.
Cell Culture. Human HCC cells Bel7402, HepG2, Huh7, and
SMMC-7721 and liver normal cells LO2 were purchased from SICB
(Shanghai Institute of Cell Biology, China). The cells were kept in
Dulbecco’s modified Eagle’s medium (DMEM) (Gibco, NY, USA)
possessing 10% heat-inactivated fetal bovine serum (FBS), strepto-
mycin, and penicillin at 100 U/mL at 37 °C in a humid atmosphere
with 5% CO2 and 95% air. Cells were routinely subcultured twice
weekly.
ASSOCIATED CONTENT
* Supporting Information
The Supporting Information is available free of charge on the
ACS Publications website at DOI: 10.1021/acs.jnat-
prod.8b00843.
■
S
1H NMR and 13NMR spectra for 12a−f (PDF)
AUTHOR INFORMATION
Corresponding Authors
*E-mail (Y. Ling): Lyyy111@sina.com.
*E-mail (H. Dai): dh123@ntu.edu.cn.
*Tel (Y. Zhang): +86-513-85051892. E-mail: zhangyanan@
gmail.com.
■
ORCID
Yong Ling: 0000-0002-3373-011X
Notes
The authors declare no competing financial interest.
MTT Assay. Briefly, 100 μL of appropriate cells (104 cells/mL)
was cultured in 96-well flat-bottom microplates in DMEM medium
overnight. The cells were treated by adding 100 μL of each compound
into the respective well for 72 h. The negative control was a 0.1%
DMSO aqueous solution. After 4 h of incubation, to each well was
added 30 μL of MTT (5 mg/mL), and the absorbance of each well
was measured using an ELISA plate reader at 570 nm. The
cytotoxicity potency of tested compounds was calculated according
to the previous literature. All the IC50 data were analyzed using
GraphPadPrism software (version 4.03).
HDAC Assay. According to the directions of the HDAC1/3/6/8
fluorimetric assay kit (FLUOR DE LYS, USA), each compound at
different concentrations was incubated with each HDAC enzyme in
the presence of enzyme substrate (Boc-Lys (Ac)-AMC) at 37 °C for
60 min.42 The reaction was quenched by the addition of the lysine
developer. After 30 min, the fluorimetric emission was measured in a
fluorescence plate reader (excitation: 355 nm, emission: 460 nm).
Each HDAC inhibitory rate was measured according to the
percentage of activity relative to the control. All the IC50 data were
analyzed using GraphPadPrism software (version 4.03).
Cell Apoptosis Analysis by Flow Cytometry Assay. A total of
10 000 Bel7402 cells were cultured overnight and incubated with 12b
(1.0 and 5.0 μM), SAHA (5.0 μM), harmine (20 μM), or vehicle for
72 h. The cells were harvested and stained with annexin-V APC and
7-AAD for 15 min. Flow cytometry analysis (Calibur, BD, USA) was
used to measure the percentage of apoptotic cells, using the annexin-V
APC signal (excitation: 633 nm, emission: 660 nm) and 7-AAD signal
(excitation: 488 nm, emission: 647 nm). The data were analyzed
through WinList 3D software (version 7.1), and the histogram was
plotted using Excel (version 2016).
Acridine Orange Staining. A total of 10 000 cells were cultured
overnight and incubated with 12b (1.0, 5.0 μM), SAHA (5.0 μM),
harmine (20 μM), or vehicle for 72h. The cells were stained with 1
mg/mL AO for 15−20 min. The AO was then removed and visualized
through a fluorescence microscope (ImageXpress Micro, MD, USA).
Western Blot Analysis. The activity of HDAC-related proteins,
autophagy-related proteins, Akt, and mTOR was analyzed by the
Western blot assay. Bel7402 cells (1.5 × 105/mL) were treated with
or without SAHA, 12b, or harmine at indicated dosages for 24 or 72
h. The cell lysates (0.05 mg/lane) were separated by SDS-PAGE
(12% gel) and transferred onto nitrocellulose membranes. When the
reaction was blocked using 5% nonfat milk, the target proteins were
then detected by different antibodies, such as anti-Bax, anti-Bcl-2,
anti-acetyl-H3, anti-acetyl-α-tubulin, anti-LC3, anti-p62, anti-Akt,
anti-phospho-Akt (Ser473), anti-mTOR, anti-phospho-mTOR, anti-
beclin-1, and anti-β-actin antibodies (Cell Signaling Technology, MA,
USA), respectively. The bound antibodies were detected by horse
radish peroxidase-conjugated second antibodies and then visualized
using the chemiluminescent reagent.
ACKNOWLEDGMENTS
■
This work was financially supported by the Project of “Jiangsu
Six Peaks of Talent” (2016-SWYY-CXTD-008 and 2014-
SWYY-044), China and Jiangsu Province Postdoctoral Science
Foundation (2018T110533, 2016M590488, and 1601136B),
the Project of “Jiangsu 333 High-Level Talents”, Applied
Research Projects of Nantong City (MS12018079 and
JC2018125), and the testing service was provided by Analysis
and Testing Center in Nantong University.
REFERENCES
■
(1) Hostettmann, K. Pure Appl. Chem. 1999, 70.
(2) Chen, G. T.; Lu, Y.; Yang, M.; Li, J. L.; Fan, B. Y. Phytother. Res.
2018, 32, 823−864.
(3) Harvey, A. L.; Young, L. C.; Viljoen, A. M.; Gericke, N. P. J.
Ethnopharmacol. 2011, 137, 1124−1129.
(4) McChesney, J. D.; Venkataraman, S. K.; Henri, J. T.
Phytochemistry 2007, 68, 2015−22.
(5) Chen, J. G.; Kensler, T. W. Chem. Res. Toxicol. 2014, 27, 3−6.
(6) McGuire, S. Adv. Nutr. 2016, 7, 418−419.
(7) Longley, D. B.; Johnston, P. G. J. Pathol. 2005, 205, 275−92.
(8) Huang, J.; Mei, H.; Tang, Z.; Li, J.; Zhang, X.; Lu, Y.; Huang, F.;
Jin, Q.; Wang, Z. J. Controlled Release 2017, 245, 1−14.
(9) Han, M. L.; Zhao, Y. F.; Tan, C. H.; Xiong, Y. J.; Wang, W. J.;
Wu, F.; Fei, Y.; Wang, L.; Liang, Z. Q. Acta Pharmacol. Sin. 2016, 37,
1606−1622.
́
́
(10) Herraiz, T.; Gonzalez, D.; Ancín-Azpilicueta, C.; Aran, V. J.;
́
Guillen, H. Food Chem. Toxicol. 2010, 48, 839−43.
(11) Moloudizargari, M.; Mikaili, P.; Aghajanshakeri, S.; Asghari, M.
H.; Shayegh, J. Pharmacogn. Rev. 2013, 7, 199−212.
(12) Lamchouri, F. Plant Sci. Today 2014, 1, 192−195.
(13) Han, X.; Zhang, J.; Guo, L.; Cao, R.; Li, Y.; Li, N.; Li, N.; Ma,
Q.; Wu, J.; Wang, Y.; Si, S. PLoS One 2012, 7, No. e46546.
(14) Herraiz, T.; Gonzalez, D.; Ancin-Azpilicueta, C.; Aran, V. J.;
Guillen, H. Food Chem. Toxicol. 2010, 48, 839−845.
(15) Chen, Q.; Chao, R.; Chen, H.; Hou, X.; Yan, H.; Zhou, S.;
Peng, W.; Xu, A. Int. J. Cancer 2005, 114, 675−82.
(16) Hamsa, T. P.; Kuttan, G. Eur. J. Pharmacol. 2010, 649, 64−73.
(17) Carvalho, A.; Chu, J.; Meinguet, C.; Kiss, R.; Vandenbussche,
G.; Masereel, B.; Wouters, J.; Kornienko, A.; Pelletier, J.; Mathieu, V.
Eur. J. Pharmacol. 2017, 805, 25−35.
(18) (a) Abbassi, R.; Johns, T. G.; Kassiou, M.; Munoz, L.
Pharmacol. Ther. 2015, 151, 87−98. (b) Ma, Y.; Wink, M. Phytother.
Res. 2010, 24, 146−9.
(19) Vishwakarma, V.; New, J.; Kumar, D.; Snyder, V.; Arnold, L.;
Nissen, E.; Hu, Q.; Cheng, N.; Miller, D.; Thomas, A. R.; Shnayder,
H
DOI: 10.1021/acs.jnatprod.8b00843
J. Nat. Prod. XXXX, XXX, XXX−XXX

Journal of Natural Products
Article
Y.; Kakarala, K.; Tsue, T. T.; Girod, D. A.; Thomas, S. M. Sci. Rep.
2018, 8, 12163.
(20) Atteya, R.; Ashour, M. E.; Ibrahim, E. E.; Farag, M. A.; El-
Khamisy, S. F. Mech. Ageing Dev. 2017, 161, 141−148.
(21) Ramsay, R. R.; Popovic-Nikolic, M. R.; Nikolic, K.; Uliassi, E.;
Bolognesi, M. L. Clin. Transl. Med. 2018, 7, 3.
(22) Petrelli, A.; Giordano, S. Curr. Med. Chem. 2008, 15, 422−432.
(23) Bose, P.; Dai, Y.; Grant, S. Pharmacol. Ther. 2014, 143, 323−
36.
(24) Qin, H. T.; Li, H. Q.; Liu, F. Expert Opin. Ther. Pat. 2017, 27,
621−636.
(25) Kim, H. Y. S.; Lee, S. M. Eur. J. Pharm. Sci. 2012, 45, 708−715.
(26) Serafim, T. L.; Carvalho, F. S.; Marques, M. P.; Calheiros, R.;
Silva, T.; Garrido, J.; Milhazes, N.; Borges, F.; Roleira, F.; Silva, E. T.;
Holy, J.; Oliveira, P. J. Chem. Res. Toxicol. 2011, 24, 763−774.
(27) Whent, M.; Huang, H.; Xie, Z.; Lutterodt, H.; Yu, L.; Fuerst, E.
P.; Morris, C. F.; Yu, L. L.; Luthria, D. J. J. Agric. Food Chem. 2012,
60, 2129−2135.
(28) Serafim, T. L.; Marques, M. P.; Borges, F.; Oliveira, P. J. J.
Theor. Exp. Pharm. 2008, 1, 3−13.
(29) Cione, E.; Tucci, P.; Senatore, V.; Perri, M.; Trombino, S.;
Iemma, F.; Picci, N.; Genchi, G. J. Bioenerg. Biomembr. 2008, 40, 19−
26.
(30) Lane, A. A.; Chabner, B. A. J. Clin. Oncol. 2009, 27, 5459−
5468.
(31) (a) Gryder, B. E. Future Med. Chem. 2012, 4, 1369−1370.
(b) Marks, P. A. Expert Opin. Invest. Drugs 2010, 19, 1049−66.
(32) Huang, M.; Liu, G. Cancer Lett. 1998, 135, 97−105.
(33) Shu, Y. J.; Bao, R. F.; Wu, X. S.; Weng, H.; Ding, Q.; Cao, Y.;
Li, M. L.; Mu, J. S.; Wu, W. G.; Ding, Q. C.; Liu, T. Y.; Jiang, L.; Hu,
Y. P.; Tan, Z. J.; Wang, P.; Liu, Y. B. Anti-Cancer Agents Med. Chem.
2014, 14, 1136−1145.
(34) Sha, M.; Ye, J.; Zhang, L. X.; Luan, Z. Y.; Chen, Y. B.; Huang, J.
X. Pharmacology 2014, 93, 39−46.
(35) Lu, J.; Qin, Q.; Zhan, L.; Liu, J.; Zhu, H.; Yang, X.; Zhang, C.;
Xu, L.; Liu, Z.; Wang, D.; Cui, H.; Meng, C.; Cai, J.; Cheng, H.; Sun,
X. J. Pharmacol. Sci. 2014, 126, 56−65.
(36) Choudhary, G. S.; Al-Harbi, S.; Almasan, A. Methods Mol. Biol.
2015, 1219, 1−9.
(37) Yu, Z.; Jie, H.; Li, Z. Med. Chem. Res. 2013, 22, 2505−2510.
(38) Sun, S. Y. Autophagy 2010, 6, 184−5.
(39) Lu, Y.; Zhang, R.; Liu, S.; Zhao, Y.; Gao, J.; Zhu, L. Eur. J.
Pharmacol. 2016, 15, 130−8.
(40) Wang, F.; Ma, H.; Liu, Z.; Huang, W.; Xu, X.; Zhang, X.
Biomed. Pharmacother. 2017, 92, 672−680.
(41) Rodon, J.; Dienstmann, R.; Serra, V.; Tabernero, J. Nat. Rev.
Clin. Oncol. 2013, 10, 143−153.
(42) Ling, Y.; Xu, C.; Luo, L.; Cao, J.; Feng, J.; Xue, Y.; Zhu, Q.; Ju,
C.; Li, F.; Zhang, Y.; Zhang, Y.; Ling, X. J. Med. Chem. 2015, 58,
9214−9227.
I
DOI: 10.1021/acs.jnatprod.8b00843
J. Nat. Prod. XXXX, XXX, XXX−XXX