β-Carboline and N-hydroxycinnamamide hybrids as anticancer agents for drug-resistant hepatocellular carcinoma

Article abstract of DOI:10.1016/j.ejmech.2019.02.054

In an effort to develop anticancer agents that may overcome drug resistance, the number one reason in caner death, we have developed a series of novel hybrids of β-carboline and N-hydroxycinnamamide as histone deacetylase (HDAC) inhibitors. Most of the hybrids 13a-p showed strong antiproliferative effects with low-micromolar IC₅₀ values against four human cancer cells. The most potent compound of series 13p exhibited high HDAC1/6 inhibitory effects, and also increased the acetylation levels of histone H3, H4 and α-tubulin. Importantly, 13p demonstrated high anticancer potency against drug-sensitive HepG2 and Bel7402 cells and drug-resistant Bel7402/5FU cells. Hybrid 13p triggered significant apoptosis by regulating apoptotic relative proteins expression in these Bel7402/5FU cells. Finally, 13p induced a substantial amount of autophagic flux activity by the accretion of the expression of LC3-II and the degeneration of expression of p62 and LC3-I in Bel7402/5FU cells. Overall, 13p is a novel β-carboline/N-hydroxycinnamamide hybrid with significant anticancer potency that warrants further evaluation for the treatment of drug-resistant hepatocellular carcinoma.

Full text of DOI:10.1016/j.ejmech.2019.02.054

Accepted Manuscript
β-Carboline and N-hydroxycinnamamide hybrids as anticancer agents for drug-
resistant hepatocellular carcinoma
Yong Ling, Wei-Jie Gao, Changchun Ling, Ji Liu, Chi Meng, Jianqiang Qian, Siqun
Liu, Huiling Gan, Hongmei Wu, Jinhua Tao, Hong Dai, Yanan Zhang
PII:
S0223-5234(19)30168-0
DOI:
https://doi.org/10.1016/j.ejmech.2019.02.054
EJMECH 11146
Reference:
To appear in: European Journal of Medicinal Chemistry
Received Date: 29 November 2018
Revised Date: 15 January 2019
Accepted Date: 17 February 2019
Please cite this article as: Y. Ling, W.-J. Gao, C. Ling, J. Liu, C. Meng, J. Qian, S. Liu, H. Gan, H. Wu,
J. Tao, H. Dai, Y. Zhang, β-Carboline and N-hydroxycinnamamide hybrids as anticancer agents for
drug-resistant hepatocellular carcinoma, European Journal of Medicinal Chemistry (2019), doi: https://
doi.org/10.1016/j.ejmech.2019.02.054.
This is a PDF file of an unedited manuscript that has been accepted for publication. As a service to
our customers we are providing this early version of the manuscript. The manuscript will undergo
copyediting, typesetting, and review of the resulting proof before it is published in its final form. Please
note that during the production process errors may be discovered which could affect the content, and all
legal disclaimers that apply to the journal pertain.

ACCEPTED MANUSCRIPT
β-Carboline and N-hydroxycinnamamide hybrids as anticancer
agents for drug-resistant hepatocellular carcinoma
Yong Ling a,b, Wei-Jie Gao a, Changchun Linga,b, Ji Liu a, Chi Meng a, Jianqiang Qiana, Siqun Liua,
Huiling Gan a, Hongmei Wu a, Jinhua Tao a, Hong Dai a,c,*, Yanan Zhang a,*
aSchool of Pharmacy and Jiangsu Province Key Laboratory for Inflammation and Molecular Drug Target, Nantong
University, Nantong 226001, PR China.
bThe Affiliated Hospital of Nantong University, Nantong University, Nantong 226001, PR China.
cCollege of Chemistry and Chemical Engineering, Nantong University, Nantong 226019, PR China
Graph Abstract

ACCEPTED MANUSCRIPT
β-Carboline and N-hydroxycinnamamide hybrids as anticancer
agents for drug-resistant hepatocellular carcinoma
Yong Ling a,b, Wei-Jie Gao a, Changchun Linga,b, Ji Liu a, Chi Meng a, Jianqiang Qiana, Siqun Liua,
Huiling Gan a, Hongmei Wu a, Jinhua Tao a, Hong Dai a,c,*, Yanan Zhang a,
aSchool of Pharmacy and Jiangsu Province Key Laboratory for Inflammation and Molecular Drug Target, Nantong
University, Nantong 226001, PR China.
bThe Affiliated Hospital of Nantong University, Nantong University, Nantong 226001, PR China.
cCollege of Chemistry and Chemical Engineering, Nantong University, Nantong 226019, PR China
Abstract: In an effort to develop anticancer agents that may overcome drug resistance, the
number one reason in caner death, we have developed a series of novel hybrids of β-carboline and
N-hydroxycinnamamide as histone deacetylase (HDAC) inhibitors. Most of the hybrids 13a-p
showed strong antiproliferative effects with low-micromolar IC50 values against four human
cancer cells. The most potent compound of the series 13p exhibited high HDAC1/6 inhibitory
effects, and also increased the acetylation levels of histone H3, H4 and α-tubulin. Importantly, 13p
demonstrated high anticancer potency against drug-sensitive HepG2 and Bel7402 cells and
drug-resistant Bel7402/5FU cells. Hybrid 13p triggered significant apoptosis by regulating
apoptotic relative proteins expression in these Bel7402/5FU cells. Finally, 13p induced a
substantial amount of autophagic flux activity by the accretion of the expression of LC3-II and the
degeneration of expression of p62 and LC3-I in Bel7402/5FU cells. Overall, 13p is a novel
β-carboline/N-hydroxycinnamamide hybrid with significant anticancer potency that warrants
further evaluation for the treatment of drug-resistant hepatocellular carcinoma.
Keywords: Histone deacetylase inhibitors; Hybrids; N-Hydroxycinnamamide; β-Carbolines; Drug
resistance.
1. Introduction
Cancer represents a major health threat in both developed and under developed countries.
According to the report of World Health Organization, an estimated 9 million patients died of
malignant tumor in the whole world in 2015 [1]. Tumor resistance has severely limited the
effectiveness of many current chemotherapeutic agents. Despite the great advances in
chemotherapy, particularly molecularly targeted therapeutics, many types of cancers that are
initially susceptible to chemotherapy can develop resistance over time through various
mechanisms. Currently 90% of failures in chemotherapy occur during the invasion and metastasis
of cancers related to drug resistance and most patients have to succumb to the recurrent
drug-resistant disease [2]. There is a pressing need to develop novel therapeutic agents to
overcome the limitations of currently available therapies, particularly the development of drug
Corresponding author. Tel.: +86-513-85051892; e-mail: zhangyanan@gmail.com, dh123@ntu.edu.cn

ACCEPTED MANUSCRIPT
resistance [3-6].
The majority of anticancer agents can trace their original or underlying synthetic design principles
to natural products [7], including many widely used antitumor drugs such as paclitaxel, teniposide,
camptothecin and vincristine [8]. β-Carbolines are a family of alkaloids found in plants and
animals, and are an important class of natural products widely distributed in nature. Many
β-carbolines, including harmine and its analogs harmaline and harmalol, have been traditionally
used for the prevention and treatment of multiple cancers or malaria [9-11]. While β-carbolines
generally show weak anti-cancer effects [12], structurally modified derivatives have been shown
to have improved antitumor activities [13]. Studies by our group and others showed that hybrid
molecules bearing a β-carboline and another antitumor entity such as a histone deacetylase
(HDAC) inhibitor may result in synergistic effects and increased anticancer efficacy [14,15].
Interestingly, harmine was also found to reduce resistance to anticancer drugs such as
camptothecin and mitoxantrone, which were mediated by breast cancer resistance protein (BCRP)
[16].
Hybrid drug molecules, designed to simultaneously modulate multiple oncogenic signaling
pathways with a single molecule, offer a promising alternate strategy that may overcome some of
the many disadvantages of single cancer drugs such as multi-drug resistance. HDACs have been a
promising candidate for hybrid agent development [17], since they are clinically validated cancer
targets and HDAC inhibitors demonstrated prominent antitumor efficacy on broad spectrum
neoplasms in preclinical and clinical studies [18,19]. For example, HDAC inhibitor belinostat
(PXD101) and cisplatin, when used together, have been found to exert synergistic cytotoxic effect
against cisplatin-resistant lung cancer cell lines, suggesting HDAC inhibitors as novel agents for
drug resistance reversal in combination chemotherapeutic regimens [20]. Therefore, introduction
of HDAC inhibitor structural moieties to β-carbolines may not only increase the potency, but also
reverse drug resistance, the number one reason in cancer death.
Fig.
1.
Structures
of
representative
N-hydroxyarylamide
based
HDAC
inhibitors
and
β-carboline/N-hydroxycinnamamide hybrids 13a-p.
A number of HDAC inhibitors have been either approved by the FDA or currently in clinical
stages, many of which have an N-hydroxycinnamamide as an important active fragment (Figure 1).
These include belinostat, panobinostat (LBH589) and pracinostat (SB939), all of which are FDA

ACCEPTED MANUSCRIPT
approved drugs for the treatment of peripheral T-cell lymphoma or multiple myeloma [21], and
dacinostat (LAQ824), which entered clinical trials [22] (Fig. 1). HDACi’s have been known to
induce autophagy and apoptosis, two forms of programed cell death that are responsible for the
turnover/degradation of organelles and proteins within cells and of cells within organisms [23].
HDACi’s commonly induce apoptosis through either the extrinsic (death receptor on cell surface)
pathway, involving ligands such as Fas or TNF-related apoptosis-inducing ligand (TRAIL) and
proteins such as FAS-associated death domain (FADD) and caspase-8, or the intrinsic
(mitochondria) pathway, involving proteins such as the Bcl-2 superfamily. Similarly, several
signaling pathways have been suggested to play important roles in regulating HDAC-induced
autophagy, including mTOR, AIF, ROS, CDKs, and HDAC1/6 [24–26]. In addition, β-carbolines
have also been reported to show antitumor activities via inducing both apoptosis and autophagy
[27-29]. Therefore, incorporation of the HDAC inhibiting N-hydroxycinnamamide moiety into the
β-carboline may lead to hybrid molecules with increased anticancer potency/efficacy. Since these
HDAC inhibitors all bear an aromatic ring-containing group on the N-hydroxycinnamamide group,
we reasoned that replacement of this aromatic group with an aromatic β-carboline may retain the
HDAC activity. In addition, the introduction of the β-carboline group with anti-tumor activities
may also render synergistic effects to these hybrids. Since the aromatic ring-containing group is at
either the meta or para position of the N-hydroxycinnamamide in these HDAC inhibitors, we will
investigate both of these substitution patterns with the β-carbolines. We herein report our progress
on the synthesis and discovery of β-carboline/N-Hydroxycinnamamide hybrids 13a-p.
2. Results and discussion
2.1 Chemistry

ACCEPTED MANUSCRIPT
Scheme 1. Preparation of compounds 13a-p: (a) malonic acid, DMF, piperidine, reflux, 4 h, 80-87%;
(b) sulfoxide chloride, methanol, 0 , 1 h, and 65
4 h, 87-91%; (c) NBS, AIBN, CCl4, rt, 2 h,
65-72%; (d) H+ or OH-, R-CHO, reflux, 2-4 h, 71-86%; (e) sulfoxide chloride, methanol, 0 , 1 h, and
65 x 4-6 h, 82-90%; (f) potassium permanganate, N,N-dimethylformamide, rt, 2-6 h, 61-76%; (g)
hydrazine monohydrate, methanol, 0 , 1 h, 84-95%; (h) NaNO2, HCl, H2O, 0 , 6 h; (i) HAc, H2O,
reflux, 6h, 56-71%; (j) 4a,b, K2CO3, CH3CN, 50 , 3-5h, 65-78%; (k) NH2OK, methanol, rt, 8-12h,
47-58%.
The preparation route of target compounds 13a-p is presented in Scheme 1. The key intermediates
4a,b were synthesized in three steps. The starting 3- and 4-methyl benzaldehydes 1a,b were
converted to cinnamic acids 2a,b in a Knoevenagel condensation reaction with the treatment of
malonic acid. Cinnamic acids 2a,b were converted to the corresponding esters 3a,b under standard
esterification conditions using MeOH and SOCl2. Bromination of 3a,b via a free radical reaction
using N-bromosuccinimide (NBS) and catalyzed by azodiisobutyronitrile (AIBN) afforded the
benzyl bromides 4a,b. Next, L-tryptophan 5 underwent a Pictet-Spengler condensation reaction
with different aldehydes to generate 6a-h, wihch were then converted to methyl esters 7a-h.
Treatment of esters 7a-h with potassium permanganate in N,N-dimethylformamide produced
intermediates 8a-h, which were then converted to β-carboline hydrazides 9a-h when treated with
hydrazine monohydrate. Treatment of 9a-h with NaNO2 accomplished the conversion of the
hydrazide group to an acyl azide group to afford intermediates 10a-h. The acyl azide in
compounds 10a-h was converted to an amino group in the presence of HAc in H2O via Curtis
rearrangement to afford compounds 11a-h. Reaction between 11a-h and methyl

ACCEPTED MANUSCRIPT
bromomethyl-cinnamate 4a,b in the presence of K2CO3 in CH3CN produced key intermediates
12a-p. Finally, the target compounds 13a-p were prepared by the reaction of intermediates 12a-p
with NH2OK in MeOH. The structure of all final compounds were confirmed using MS, 1H NMR,
13C NMR and HRMS. All target compounds had purity of > 95% as determined using HPLC.
3. Results and discussion
3.1. HDAC1 inhibitory activity
It was our hypothesis that replacement of the aromatic ring on N-hydroxycinnamamide of
HDAC inhibitors such as belinostat, panobinostat and pracinostat with a β-carboline subunit
would retain their activity. To confirm this was indeed the case, all target compounds 13a-p were
first screened for their HDAC1 inhibitory effect. HDAC1 is a key HDAC subtype and has been
indicated in cancer cell proliferation, differentiation, and tissue malignant proliferation. As seen in
Table 1, the β-carboline, harmine, showed little HDAC1 inhibitory activity (IC50 > 1000 nM),
consistent with previous reports. However, all of compounds 13a-p exhibited significant HDAC
inhibitory effects with nanomolar IC50 values.
The structure-activity relationship of compounds 13a-p clearly revealed that the HDAC1
inhibitory activities of the series are clearly dependent on the substituents at the C1 position.
Compounds 13c,d with a methyl group or 13a,b with a hydrogen displayed weak HDAC1
inhibition. However, the potencies were enhanced when aryl groups were introduced at the C1
position. In particular, compounds with a phenyl group bearing electron-donating substituents
such as a methoxyl group (13g-j) or two or more methoxyl groups (13m-p) displayed greater
HDAC1 inhibitory potencies than the rest compounds. Compound 13p was the most potent
compound of the series and its IC50 value (1.3 nM against HDAC1) was >100-fold lower than the
FDA approved HDAC inhibitor SAHA.
3.2. Antiproliferative activities of target compounds
Given the apparent HDAC inhibitory activities of these compounds, some of which are even
more potent than SAHA, these target compounds would possess potent antitumor activities, as
expected with HDAC inhibitors. To examine their inhibitory activitiy on cancer cell growth, all
target compounds 13a-p were first screened in the MTT assay against human colon cancer cells
(HCT116), human hepatocellular carcinoma cells (HepG2 and SMMC-7721), and human lung
cancer cells (H1299). Their IC50 values are summarized in Table 1. Harmine and SAHA were also
tested for comparison.
Table 1. IC50 values of hybrids 13a-p against four human cancer cell lines and HDAC1a
In vitro antiproliferative activity (IC50a, µM)
HDAC1
Compd
R
m or p
IC50 (nM)
142 ± 18
>1000
SUMM-7721
5.23 ± 0.48
47.6 ± 5.12
>12.5
Hep G2
HCT116
H1299
SAHA
Harmine
13a
/
/
/
5.94 ± 0.73 4.97 ± 0.56 7.02 ± 0.65
/
H
53.6 ± 5.50 43.8 ± 3.85
NDb
>12.5
>12.5
>12.5
p
197 ± 18
161 ± 15
122 ± 14
93 ± 9
>12.5
>12.5
>12.5
>12.5
>12.5
H
m
p
>12.5
13b
Me
10.6 ± 0.92
7.35 ± 0.80
2.99 ± 0.41
11.3 ± 1.07
13c
Me
m
p
8.81 ± 1.02 9.12 ± 0.76 8.93 ± 0.61
3.46 ± 0.48 3.31 ± 0.29 2.95 ± 0.36
13d
4-Me-Ph
65 ± 9
13e

ACCEPTED MANUSCRIPT
4-Me-Ph
3-MeO-Ph
m
p
81 ± 10
1.8 ± 0.3
5.2 ± 0.9
26 ± 4
5.19 ± 0.70
1.01 ± 0.14
2.42 ± 0.31
3.23 ± 0.39
2.51 ± 0.30
2.18 ± 0.21
1.75 ± 0.15
5.46 ± 0.49
4.10 ± 0.56
1.59 ± 0.14
1.01 ± 0.10
6.17 ± 0.52 4.99 ± 0.58 6.32 ± 0.71
0.72 ± 0.10 1.36 ± 0.18 1.25 ± 0.21
1.23 ± 0.17 1.96 ± 0.30 1.39 ± 0.24
2.80 ± 0.33 4.29 ± 0.46 2.78 ± 0.32
1.96 ± 0.24 2.47 ± 0.28 3.03 ± 0.42
1.62 ± 0.15 2.95 ± 0.32 2.46 ± 0.30
1.27 ± 0.16 1.68 ± 0.20 2.22 ± 0.26
3.85 ± 0.43 4.91 ± 0.61 7.06 ± 0.58
3.41 ± 0.47 4.33 ± 0.52 6.11 ± 0.53
1.55 ± 0.18 1.23 ± 0.14 1.90 ± 0.22
0.41 ± 0.06 0.87 ± 0.11 0.69 ± 0.10
13f
13g
13h
13i
3-MeO-Ph
m
p
4-MeO-Ph
4-MeO-Ph
m
p
9.7 ± 1
20 ± 3
13j
4-NO2-Ph
13k
13l
4-NO2-Ph
m
p
13 ± 2
3,4-(MeO)2-Ph
3,4-(MeO)2-Ph
3,4,5-(MeO)3-Ph
3,4,5-(MeO)3-Ph
24 ± 4
13m
13n
13o
13p
m
p
9.3 ± 2
5.6 ± 0.7
1.3 ± 0.2
m
a Related data appeared as the means ± SD of three separated assays. b ND: not detected.
As shown in Table 1, most hybrids showed good antiproliferative potencies in all 4 cells, with
IC50’s in the low micromolar range. Among them, compounds 13e, 13g-l, and 13o,p (IC50
=
0.41-4.29 µM) demonstrated better antiproliferative potencies than SAHA (IC50 = 4.97-7.02 µM).
Consistent with literature reports, harmine displayed weak anti-cancer potency. These results
indicate that the designed structural hybridization resulted in molecules with good to excellent
antiproliferative properties. Furthermore, the relative spatial arrangement of the substituted
β-carboline moiety and N-hydroxycinnamamide had little impact on the potency, with the meta
analogs generally exerting a slightly stronger growth inhibitory activity than those with the para
position substitutions (e.g. 13m vs. 13n, and 13o vs. 13p).
In comparing the enzymatic activities with cell viabilities, it is apparent that the relative
HDAC1 potencies of these hybrids matched well with their cell based antiproliferation. These
results suggest that the antiproliferative properties of this series of hybrids result largely from their
HDAC inhibitory activities. Given the strong inhibitory activity of 13p against tumor cells in vitro,
we further examined its inhibitory effects on the growth of normal liver cells LO2. It was found
that 13p had weak inhibitory effect on non-tumor LO2 cells with an IC50 value of 2.83 µM, which
was nearly seven-fold weaker than the HepG2 cells (IC50 = 0.41 µM).
3.3. 13p displays antiproliferative effects against 5-FU drug sensitive and resistance cells
Drug resistance is a leading cause for the failure of many anticancer chemotherapeutic agents
[30,31]. To investigate whether these hybrids with potent antiproliferative activities against 4
cancer cells also affect drug resistance, the most potent compounds 13g,h, 13k,l, and 13o,p were
further assayed for their activity against drug-sensitive hepatocellular carcinoma (HCC) Bel7402
and the 5-fluorouracil-resistant HCC Bel7402/5-FU cells in the MTT assay. These Bel7402/5-FU
cells have been shown to display multi-drug resistance with cross-resistance to many drugs
including adriamycin, vincristine, and oxaliplatin [32]. As shown in Table 2, all these compounds
displayed significant antiproliferative effects in both cells, with IC50 values in the low micromolar
levels and lower than both SAHA and 5-FU. Notably, 13p demonstrated the greatest
anti-proliferative effects with IC50 values of 0.85 and 2.09 µM against the drug-sensitive Bel7402
and drug-resistant Bel7402/5-FU cell lines. There values were nearly five- to six-fold better than
SAHA (IC50 = 4.72-9.83 µM) and eighteen- to thirty-fold more potent than 5-FU (IC50 = 15.6-61.7
µM). In comparison, compound 12p, an intermediate of 13p, exhibited weaker antiproliferative

ACCEPTED MANUSCRIPT
activity than 13p in both cells, confirming the importance of the hydroxamic acid.
5-FU is one of the most widely used antitumor drugs against colon, liver, and lung cancers.
While the mechanisms of 5-FU resistance have not been fully understood, a number of
mechanisms have been proposed such as overexpression of target enzyme thymidylate synthetase
(TS), high metabolism of 5-FU through the overexpression of dihydropyrimidine dehydrogenase
(DPD), and increase of the efflux of 5-FU from cancer cells through ATP-binding cassette (ABC)
transporters [33–35]. While further studies are clearly needed, the significant potency displayed by
these hybrids in drug-sensitive and drug-resistant cells probably resulted from the synergistic
effects between the HDAC inhibitor and the β-carboline component. In fact, a number of studies
have demonstrated that HDACi’s enhance or potentiate 5-FU cytotoxicity, via mechanisms
including down-regulation of thymidylate synthase in human cancer cells, or upregulation of
major histocompatibility complex class II and p21 genes and activates caspase-3/7 [33-35]. In
addition, CUDC-907, a dual HDAC and PI3K inhibitor enhanced the efficacy of 5-FU when used
in combination in colorectal cancer cells [36], similar to our results described here.
Table 2. IC50 values of compounds 13g,h, 13k,l, and 13o,p against drug-sensitive HCC Bel7402 and drug-resistant
HCC Bel7402/5-FU cells a
IC50a, µM
Compd.
Bel7402
Bel7402/5-FU
9.83 ± 0.89
61.7 ± 8.45
2.17 ± 0.26
2.90 ± 0.35
4.38 ± 0.56
3.06 ± 0.29
3.66 ± 0.48
2.09 ± 0.30
16.2 ± 1.95
SAHA
5-FU
13g
4.72 ± 0.61
15.6 ± 2.03
1.03 ± 0.12
2.15 ± 0.23
3.24 ± 0.35
2.71 ± 0.40
1.97 ± 0.26
0.85 ± 0.13
12.7 ± 1.53
13h
13k
13l
13o
13p
12p
aThese data are from three independent experiments, and are presented as the mean ± SD.
3.4. HDAC subtype selectivity
In addition to HDAC1, several other members of the HDAC family have been indicated in
cancer, including HDAC3, HDAC6, and HDAC8. In fact, it has been suggested that the potential
side effects of HDAC inhibitors depend largely on the subtype selectivity of the compounds [37].
Therefore, the compounds that showed the best potency for HDAC1 inhibition, including 13g, 13h,
13o, and 13p, were further evaluated for their inhibitory activities against HDAC3/6/8 (Table 3).
This was accomplished by measuring the fluorescent-based enzyme activity of human
recombinant HDAC3/6/8 enzymes. These four compounds possessed high inhibitory effects
against HDAC6 with IC50 values of 2.8-7.3 nM, which were significantly lower than SAHA.
However, these compounds showed moderate HDAC3 inhibition and weak HDAC8 inhibition.
Notably, the most potent compound of the series 13p exhibited high HDAC1/6 inhibitory effects.
This selectivity of 13p for HDAC1/6 may result from its active fragment N-hydroxycinnamamide,
which not only can form bi-chelation with the active Zn2+ binding site with its hydroxamic acid
group, but also can possibly form sandwich-like π−π interactions by inserting its vinyl benzene

ACCEPTED MANUSCRIPT
group into the two parallel phenylalanine residues of HDAC1/6, as previously reported [38]. In
addition, the cap group (β-carboline) may also contribute to the observed HDAC1/6 selectivity.
Due to the flexibility of loops in the cap region of HDAC1 and HDAC6, many N-containing
polyaromatics as cap groups, such as quinazolines and β-carbolines, can comfortably occupy the
surface groove and interact with the amino acid residues present at the surface of the HDACs [39].
Table 3. IC50 values of hybrids 13g, 13h, 13o, and 13p against HDAC1, HDAC3, HDAC6, and HDAC8a
IC50 (nM)a
Compd.
HDAC1
142 ± 18
1.8 ± 0.3
5.2 ± 0.9
5.6 ± 0.7
1.3 ± 0.2
HDAC3
153 ± 17
76 ± 9
HDAC6
65 ± 7
HDAC8
425 ± 51
617 ± 55
>1000
SAHA
13g
2.8 ± 0.3
4.5 ± 0.6
7.3 ± 0.9
3.1 ± 0.5
83 ± 9
13h
185 ± 21
146 ± 18
>1000
13o
>1000
13p
a These data are from three independent experiments, and are expressed as the mean ± SD.
3.5. Acetylated histone H3/4 and α-tubulin induced by 13p.
To confirm that these hybrids induce acetylation of histones at the cellular level, the most potent
compound of the series, 13p, was analyzed in western blotting assays for its effect on the
expression level of the Ac-H3, Ac-H4, and Ac-α-tubulin. β-Actin was used as a control (Fig. 2).
Drug resistant HCC Bel7402/5-FU cells were incubated for 72h with the vehicle, SAHA (5.0 µM),
or 13p (0.8, 2.0, and 5.0 µM). The levels of Ac-H3/4, and Ac-α-tubulin relative to β-actin were
tested by densimetric scanning.
It is well known that α-tubulin is the target of HDAC6 and histones H3 and H4 are the targets of
HDAC1 and HDAC2. Compounds 13p significantly enhanced the levels of Ac-H3 and Ac-H4, in
agreement with its HDAC1 activity in the enzymatic assay. In addition, 13p also increased the
expression of Ac-α-tubulin, consistent with the observed high 6 activities. These results also
suggest that 13p was cell permeable. Consistent with the above HDAC isoform fluorimetric assay
results, most of the 13p treated groups (2.0 and 5.0 µM) showed higher levels of Ac-H3, Ac-H4,
and Ac-α-tubulin, than the 5.0 µM SAHA treated group.
Fig. 2. Treatment with 13p enhanced histone H3/4 and α-tubulin acetylation in vitro. (A) Western blotting analysis
in Bel7402/5-FU cells treated with 13p and SAHA. (B) Quantitative analysis. These data are from three separate
experiments, and are presented as means ± SD. *P< 0.01 vs control.
3.6. Tumor cell apoptosis induction of 13p

ACCEPTED MANUSCRIPT
Since it displayed selective HDAC1/6 inhibitory effects and remarkable HCC cell
antiproliferative activities, we next investigated whether 13p leads to enhanced cancer cell
apoptosis, a common mechanism for antiproliferation. Thus, Bel7402/5-FU cells were incubated
for 72h with of 13p (2.0 or 5.0 µM), SAHA, or harmine, respectively, and then screened with
FITC-Annexin V/PI staining and flow cytometry to analysize the percentages of apoptotic cells. In
Fig. 3, the percentages of apoptotic Bel7402/5-FU cells (53.42% for 2.0 µM and 71.05% for 5.0
µM) increased with the increasing concentrations of 13p, suggested increased apoptosis. These
values were higher than both the SAHA (30.78% for 5.0 µM) and harmine (23.36% for 20 µM)
groups, agreeing with above results.
Fig. 3. Compound 13p induces Bel7402/5-FU cell apoptosis in vitro. (A) Flow cytometry analysis of
Bel7402/5-FU cells treated 13p, SAHA, or harmine, respectively, and stained with FITC-Annexin V/PI. (B)
Quantitative analysis of percentages of apoptotic cells. Related data appeared as the means ± SD of three separated
assays. *P< 0.01 vs control.
We next determined the expression of several key apoptotic proteins, including Bax and Bcl-2,
and the cleavage states of caspase-3 and PARP (apoptosis markers), upon 13p treatment (Fig. 4).
Thus, Western blotting assay was performed using subconfluent Bel7402/5-FU HCC cells treated
with or without 13p, SAHA, or harmine for 72h. The levels of protein expression were then
detected using specific antibodies. As depicted in Fig. 4, treatment with 13p markedly increased
the expression of pro-apoptotic Bax while reducing the expression of the antiapoptotic Bcl-2
protein. Importantly, 13p led to cleavage of PARP and caspase-3 to a greater degree than the
control group, and also greeater than the SAHA group. Together, these results indicate that 13p
could significantly promote cancer cell apoptosis in drug-resistant Bel7402/5-FU cells.

ACCEPTED MANUSCRIPT
Fig. 4. (A) The levels of Bax, Bcl-2, cleaved Caspase 3, PARP and β-actin in Bel7402/5-FU cells treated with 13p,
SAHA, or harmine in Western blotting assay. (B) Quantitative analysis of the relative levels of each protein (Bax,
Bcl-2, cleaved-caspase 3, and cleaved-PARP) compared to β-actin. Related data appeared as the means ± SD of
three separated assays. *P < 0.01 vs control.
3.7. Compound 13p induces DNA damage
β-Carbolines have been reported to bind and induce DNA damage, thanks to the planar skeleton
[40]. We therefore investigated DNA damage induced by 13p in Bel7402/5-FU cells, which may
contribute to the anti-cancer potency of these β-carboline/N-hydroxycinnamamide hybrids.
Histone H2AX phosphorylation was used as the DNA damage marker. Thus, Bel7402/5-FU cells
were treated with vehicle or 13p for 72 h and their lysates were analyzed using specific antibodies.
H2AX (S139ph) levels were determined using western blotting analysis. Harmine and SAHA
were also tested in parallel. As shown in Fig. 5, 13p dose-dependently enhanced the levels of
phosphorylated H2AX in Bel7402/5-FU cells, which was significantly better than both the
harmine and SAHA group. Therefore, these results clearly support that the hybrids are involved in
DNA damage.
Fig. 5. DNA damage in western blotting assay. (A) H2AX levels, relative to β-Actin, in Bel7402/5-FU cells treated
with 13p, SAHA, or harmine. (B) Quantitative analysis. The data of three separate experiments are presented as

ACCEPTED MANUSCRIPT
means ± SD from. *P < 0.01 vs control.
3.8. Compound 13p induces autophagy in Bel7402/5-FU cells
Autophagy is an intracellular catabolic process that targets damaged and superfluous cellular
proteins, organelles, and other cytoplasmic components to maintain metabolism and homeostasis
[41]. The role of autophagy in cancer is context dependent, but studies in genetically engineered
mouse models suggest that autophagy can constrain tumor initiation by regulating DNA damage
and oxidative stress [42-44]. Autophagy can be induced by numerous cytotoxic compounds. To
examine whether 13p affects autophagy in Bel7402/5-FU cells, we used acridine orange (AO)
staining to detect autophagic formation using fluorescence microscopy. As shown in Fig. 6, green
fluorescence was primarily emitted in control cells with weak red fluorescence, highlighting a low
autophagic induction. By contrast, cells treated with 13p displayed a dose-dependent rise in red
fluorescence, while 13p sharply decreased the green fluorescence intensity. These results indicate
a substantial amount of the autophagosome accumulations in 13p–treated groups in Bel7402/5-FU
cells, which was more than the harmine- and SAHA-treated groups.
Fig. 6. Detection of autophagic vacuoles in Bel7402/5-FU cells after various treatments for 24 h; (a) control cells;
(b) harmine for 20 µM, (c) SAHA for 5.0 µM and (d) 13p-treated cells (2.0 and 5.0 µM). The level of acridine
orange staining was determined by fluorescence microscopy. Representative mages were captured from three
independent experiments under 400 × magnification.
The induction of autophagy was further studied by assessing the levels of key
autophagy-associated proteins. The transformation of LC3-I into LC3-II is a key step in autophagy
and the number of autophagosomes correlates to the number of LC3-II puncta. We therefore
examined the expression of autophagy-associated proteins LC3-I, LC3-II and p62, in response to
13p treatment (Fig. 7). Thus, Bel7402/5-FU cells were incubated with, or without 13p, harmine,
and SAHA at the indicated doses for 72h and the levels of protein were analyzed using specific
antibodies. Western blotting assay showed that 13p markedly enhanced the expression of LC3-II
in a dose-dependent manner, which were better than that in harmine- and SAHA-treated groups
(Fig. 7B), while expression of p62 decreased after treatment with different concentrations of 13p
for 72h.

ACCEPTED MANUSCRIPT
Fig. 7. (A) The expression of autophagy-related proteins LC3 and p62 as analyzed in western blotting assay in
Bel7402/5-FU cells treated with 13p, SAHA, or harmine. (B) Quantitative analysis of the relative levels of LC3
and p62 compared to control β-actin. Related data appeared as the means ± SD of three separated assays. *P < 0.01
vs respective control.
4. Conclusions
In summary, we have designed and prepared a series of novel hybrids based on β-carboline
and N-hydroxycinnamamide 13a-p, and evaluated their in vitro anticancer activities and
anticancer mechanisms in a battery of assays. Most of these compounds had enhanced inhibitory
potency against HDAC1 and also exerted selective HDAC6 inhibition. They also displayed low
micromolar antiproliferative potencies against 4 different human cancer cells, including 13p
which exhibited strong and selective antiproliferative activities against both drug-sensitive HepG2,
Bel7402 and drug-resistant Bel7402/5FU cells. Further, 13p caused accumulation of acetylated
histones and acetylated α-tubulin, confirming their HDAC inhibitory activities. In addition, 13p
also displayed greater potency in inducing Bel7402/5FU cell apoptosis than SAHA and harmine
by enhancing the expression of cleaved caspase-3 and PARP proteins. In addition, 13p induced a
substantial amount of autophagic flux activity in Bel7402/5FU cells by up-regulation of LC3-II
proteins and down-regulation of p62 and LC3-I, more significant than both the harmine- and
SAHA-treated groups. In summary, these results suggest that the novel
β-carboline/N-hydroxycinnamamide hybrids described here may be potential candidates for the
development of novel antitumor agents in the treatment of human cancer, especially those with
multi-drug resistance.
5. Experimental protocols
5.1. Chemical analysis
All compounds were synthesized and purified by column chromatography containing silica gel
(300–400 mesh) or by recrystallization. TLC analysis was performed by silica gel GF254 plates

ACCEPTED MANUSCRIPT
1
13
(250 mm; Qingdao O. C. C., China). H NMR and C NMR spectra were collected on a Bruker
AV 400M spectrometer in CDCl3 or DMSO-d6 and TMS was used as the internal standard. Mass
Spectra was recorded on a Mariner Mass Spectrum (ESI). High resolution mass spectrometry
(HRMS) were recorded with Agilent technologies LC/MSD TOF. Compounds 4 were prepared
according to the procedures described in literatures [45]. Compounds 6-11 were prepared
following the procedures in literature [14,46].
5.2. General preparation of 12a-p.
To a solution of compound 4 (4.5 mmol) in 50 ml acetonitrile, was added 11 (8.2 mmol),
K2CO3 (6.0 mmol), and KI (0.7 mmol) at 50 for 8h. After the reaction completed, the reaction
mixture was concentrated, and the residue was gathered through extraction and concentration, and
further purified through quick column chromatography to yield compounds 12a-p.
Compound 12a: Yield 71%, MS (ESI) m/z = 358.2 [M+H]+. Analytical data for 12a: Yellow
solid;1H NMR (d6-DMSO, 400 MHz) δ 7.91 (t, 1H, Ar-H), 7.66 (d, J = 7.4 Hz, 4H, Ar-H), 7.61
(dd, J = 16.0Hz,1H, CH=CH), 7.47 - 7.38 (m, 1H, Ar-H), 7.35 (d, J = 8.2 Hz, 1H, Ar-H), 7.10 –
6.99 (m, 2H, Ar-H), 6.59 (d, J = 16.0 Hz, 1H, CH=CH), 4.88 (s, 2H, CH2), 3.71 (s, 3H, -OCH3),
2.66 (m, 3H, CH3).
Compound 12b: Yield 68%, MS (ESI) m/z = 358.2 [M+H]+. 1H NMR (d6-DMSO, 400 MHz) δ
10.95 (s, 1H, NH), 7.90 (t, 1H, Ar-H), 7.64 (d, J = 7.4 Hz, 4H, Ar-H), 7.61 (dd, J = 15.2Hz,1H,
CH=CH), 7.47 – 7.38 (m, 1H, Ar-H), 7.35 (d, J = 8.2 Hz, 1H, Ar-H), 7.18 – 6.96 (m, 2H, Ar-H),
6.59 (d, J = 15.2Hz, 1H, CH=CH), 4.86 (s, 2H, CH2), 3.69 (s, 3H, -OCH3), 2.68 (m, 3H, CH3).
Compound 12c: Yield 65%, MS (ESI) m/z = 372.2 [M+H]+. 1H NMR (d6-DMSO, 400 MHz) δ:
10.97 (s, 1H, NH), 7.93-7.95 (d, 1H, J = 8 Hz, Ar-H), 7.61-7.67 (m, 4H, NH, Ar-H), 7.34-7.43 (m,
3H, CH=, Ar-H), 7.05-7.06 (m, 2H, Ar-H), 6.57-6.61 (d, 1H, J = 16.0 Hz, CH=), 4.88 (s, 2H, CH2),
3.71 (s, 3H, OCH3), 2.63 (s, 3H, CH3).
Compound 12d: Yield 68%, MS (ESI) m/z = 372.2 [M+H]+. 1H NMR (d6-DMSO, 400 MHz) δ:
10.97 (s, 1H, NH), 7.94-7.96 (d, 1H, J = 8.0 Hz, Ar-H), 7.35-7.64 (m, 7H, CH=, NH, Ar-H),
7.02-7.06 (m, 2H, Ar-H), 6.52-6.56 (d, 1H, J = 16.0 Hz, CH=), 4.90 (s, 2H, CH2), 3.70 (s, 3H,
OCH3), 2.65 (s, 3H, CH3).
Compound 12e: Yield 60%, MS (ESI) m/z = 448.2 [M+H]+. 1H NMR (d6-DMSO, 400 MHz) δ:
10.77 (s, 1H, NH), 8.03 (d, 1H, J = 8.0 Hz, Ar-H), 7.85-7.87 (m, 2H, Ar-H), 7.60-7.68 (m, 3H,
NH, Ar-H), 7.34-7.49 (m, 6H, CH=, Ar-H), 7.09-7.11 (m, 2H, Ar-H), 6.60 (d, 1H, J = 16.0 Hz,
CH=), 4.59-4.61 (d, 2H, J = 8.0 Hz, CH2), 3.70 (s, 3H, OCH3), 2.40 (s, 3H, CH3).
Compound 12f: Yield 61%, MS (ESI) m/z = 448.2 [M+H]+. 1H NMR (d6-DMSO, 400 MHz) δ:
10.89 (s, 1H, NH), 8.01-8.03 (d, 1H, J = 8 Hz, Ar-H), 7.86-7.89 (m, 2H, Ar-H), 7.56-7.70 (m, 2H,
NH, Ar-H), 7.32-7.48 (m, 8H, CH=, Ar-H), 7.09-7.11 (m, 1H, Ar-H), 6.51-6.55 (d, 1H, J = 16.0
Hz, CH=), 4.98 (s, 2H, CH2), 3.70 (s, 3H, OCH3), 2.40 (s, 3H, CH3).
Compound 12g: Yield 65%, MS (ESI) m/z = 464.2 [M+H]+. 1H NMR (d6-DMSO, 400 MHz) δ:
10.93 (s, 1H, NH), 8.03 (d, 1H, J = 8.0 Hz, Ar-H), 7.61-7.68 (m, 5H, NH, Ar-H), 7.35-7.55 (m,

ACCEPTED MANUSCRIPT
6H, CH=, Ar-H), 6.99-7.10 (m, 2H, Ar-H), 6.56-6.60 (m, 1H, CH=), 4.97 (s, 2H, CH2), 3.70-3.73
(m, 6H, 2 × OCH3).
Compound 12h: Yield 64%. MS (ESI) m/z = 464.2 [M+H]+. 1H NMR (d6-DMSO, 400 MHz) δ:
10.93 (s, 1H, NH), 8.03 (d, 1H, J = 8 Hz, Ar-H), 7.55-8.05 (m, 6H, NH, Ar-H), 7.34-7.50 (m, 5H,
CH=, Ar-H), 6.99-7.09 (m, 2H, Ar-H), 6.51-6.55 (d, 1H, J = 16.0 Hz, CH=), 4.97 (s, 2H, CH2),
3.70-3.73 (m, 6H, 2 × OCH3).
Compound 12i: Yield 68%. MS (ESI) m/z = 464.2 [M+H]+. 1H NMR (d6-DMSO, 400 MHz) δ:
10.81 (s, 1H, NH), 8.04 (d, 1H, J = 8.0 Hz, Ar-H), 7.61-7.67 (m, 3H, NH, Ar-H), 7.42-7.56 (m,
7H, CH=, Ar-H), 6.99-7.17 (m, 3H, Ar-H), 6.73 (m, 1H, Ar-H), 6.58 (d, 1H, J = 16.0 Hz, CH=),
4.61 (d, 2H, J = 4.0 Hz, CH2), 3.71-3.81 (m, 6H, 2 × OCH3).
Compound 12j: Yield 65%. MS (ESI) m/z = 464.2 [M+H]+. 1H NMR (d6-DMSO, 400 MHz) δ:
10.94 (s, 1H, NH), 8.04 (d, 1H, J = 8 Hz, Ar-H), 7.57-7.70 (m, 5H, NH, Ar-H), 7.34-7.51 (m, 6H,
CH=, Ar-H), 7.00-7.09 (m, 2H, Ar-H), 6.51-6.55 (m, 1H, CH=), 4.99 (s, 2H, CH2), 3.70-3.75 (m,
6H, 2 × OCH3).
Compound 12k: Yield 59%. MS (ESI) m/z = 479.2 [M+H]+. 1H NMR (d6-DMSO, 400 MHz) δ:
11.16 (s, 1H, NH), 8.07-8.39 (m, 5H, Ar-H), 7.61-7.69 (m, 4H, NH, Ar-H), 7.30-7.51 (m, 5H,
CH=, Ar-H), 7.11-7.15 (m, 1H, Ar-H), 6.56-6.61 (m, 1H, CH=), 4.99 (s, 2H, CH2), 3.70-3.76 (m,
3H, OCH3).
Compound 12l: Yield 65%. MS (ESI) m/z = 479.2 [M+H]+. 1H NMR (d6-DMSO, 400 MHz) δ:
11.16 (s, 1H, NH), 8.08-8.39 (m, 4H, Ar-H), 7.32-7.86 (m, 9H, CH=, NH, Ar-H), 7.11-7.15 (m, 1H,
Ar-H), 6.49-6.53 (d, 1H, J = 16.0 Hz, CH=), 5.01 (s, 2H, CH2), 3.70-3.72 (s, 3H, OCH3).
Compound 12m: Yield 63%. MS (ESI) m/z = 494.2 [M+H]+. 1H NMR (d6-DMSO, 400 MHz) δ:
10.90 (s, 1H, NH), 8.02 (d, 1H, J = 4 Hz, Ar-H), 7.49-7.73 (m, 5H, NH, Ar-H), 7.38-7.45 (m, 4H,
CH=, Ar-H), 7.29 (s, 1H, Ar-H), 7.06-7.10 (m, 2H, Ar-H), 6.56-6.60 (d, 1H, J = 16.0 Hz, CH=),
4.97 (s, 2H, CH2), 3.66-3.83 (m, 9H, 3 × OCH3).
Compound 12n: Yield 65%. MS (ESI) m/z = 494.2 [M+H]+. 1H NMR (d6-DMSO, 400 MHz) δ:
10.91 (s, 1H, NH), 8.03 (d, 1H, J = 4 Hz, Ar-H), 7.57-7.70 (m, 4H, NH, Ar-H), 7.31-7.49 (m, 6H,
CH=, Ar-H), 7.09-7.12 (m, 2H, Ar-H), 6.51-6.55 (d, 1H, J = 16.0 Hz, CH=), 5.00 (s, 2H, CH2),
3.68-3.86 (m, 9H, 3 × OCH3).
Compound 12o: Yield 62%. MS (ESI) m/z = 524.2 [M+H]+. 1H NMR (d6-DMSO, 400 MHz) δ:
10.95 (s, 1H, NH), 8.03-8.05 (d, 1H, J = 8 Hz, Ar-H), 7.58-7.69 (m, 4H, NH, Ar-H), 7.35-7.50 (m,
5H, CH=, Ar-H), 7.07-7.11 (m, 2H, Ar-H), 6.57-6.61 (d, 1H, J = 16.0 Hz, CH=), 4.98 (s, 2H, CH2),
3.71-3.76 (m, 12H, OCH3).
Compound 12p: Yield 67%. MS (ESI) m/z = 524.2 [M+H]+. 1H NMR (d6-DMSO, 400 MHz) δ:
10.95 (s, 1H, NH), 8.05-8.07 (d, 1H, J = 8 Hz, Ar-H), 7.57-7.70 (m, 4H, NH, Ar-H), 7.35-7.50 (m,

ACCEPTED MANUSCRIPT
5H, CH=, Ar-H), 7.08-7.13 (m, 2H, Ar-H), 6.51-6.55 (d, 1H, J = 16.0 Hz, CH=), 5.00 (s, 2H, CH2),
3.70-3.83 (m, 12H, OCH3).
5.3. General preparation of 13a-p.
A solution of NH2OK (0.25 g, 10 mmol) in 6 mL of anhydrous methanol was added to a
solution of 12 (1 mmol) in 3 mL of anhydrous methanol. The mixture was stirred for 12 h at room
temperature. Upon completion, the mixture was condensed under vacuum, and the residue was
neutralized with 2 N HCl to pH = 7, and then concentrated. The residue was purified by column
chromatography to afford target compounds 13a-p.
Compound 13a: Yield 51%, a pale yellow solid. Purity: 97.9% (by HPLC). MS (ESI) m/z =
359.2 [M+H]+; 1H NMR (d6-DMSO, 400 MHz) δ: 10.90 (s, 1H, NH), 10.33 (s, 1H, NH), 8.86 (s,
1H, Ar-H), 8.59 (s, 1H, Ar-H), 7.94 (d, 1H, J = 8.0 Hz, Ar-H), 7.69-7.73 (m, 4H, NH, Ar-H),
7.45-7.51 (m, 4H, CH=, Ar-H), 7.06 (m, 1H, Ar-H), 6.57 (d, 1H, J = 16.0 Hz, CH=), 4.87 (d, 2H,
J = 8.0 Hz, CH2). 13C NMR (d6-DMSO, 100 MHz) δ: 167.2, 151.6, 144.8, 143.2, 139.4, 133.9,
132.9, 131.4, 129.3, 128.9, 128.3, 127.9, 126.7, 122.1, 121.5, 118.5, 117.6, 112.0, 45.5. HRMS
(ESI): m/z calcd for C21H19N4O2: 359.1508; found: 359.1522.
Compound 13b: Yield 47%, a pale yellow solid. Purity: 98.3% (by HPLC). MS (ESI) m/z =
359.2 [M+H]+; 1H NMR (d6-DMSO, 400 MHz) δ: 11.78 (s, 1H, NH), 10.36 (s, 1H, NH), 8.75 (s,
1H, Ar-H), 8.50 (s, 1H, Ar-H), 7.93 (d, 1H, J = 8.0 Hz, Ar-H), 7.63-7.70 (m, 3H, NH, Ar-H),
7.34-7.49 (m, 5H, CH=, Ar-H), 7.05-7.08 (m, 1H, Ar-H), 6.58 (d, 1H, J = 16.0 Hz, CH=), 4.88 (m,
2H, CH2). 13C NMR (d6-DMSO, 100 MHz) δ: 166.7, 151.6, 145.0, 141.2, 139.2, 134.4, 133.2,
131.4, 130.0, 129.3, 129.2, 128.1, 127.8, 127.2, 122.0, 120.7, 118.4, 118.1, 114.4, 112.0, 45.6.
HRMS (ESI): m/z calcd for C21H19N4O2: 359.1508; found: 359.1499.
Compound 13c: Yield 55%, a pale yellow solid. Purity: 98.5% (by HPLC). MS (ESI) m/z =
373.2 [M+H]+; 1H NMR (d6-DMSO, 400 MHz) δ: 10.97 (s, 1H, NH), 7.94 (d, 1H, J = 8.0 Hz,
Ar-H), 7.61-7.67 (m, 4H, NH, Ar-H), 7.34-7.43 (m, 3H, CH=, Ar-H), 7.05-7.06 (m, 2H, Ar-H),
6.59 (d, 1H, J = 16.0 Hz, CH=), 4.88 (m, 2H, CH2), 3.71 (s, 3H, OCH3), 2.63 (s, 3H, CH3). 13C
NMR (d6-DMSO, 100 MHz) δ: 167.2, 151.6, 144.8, 143.2, 142.1, 139.3, 133.0, 131.4, 129.3,
128.9, 128.3, 128.1, 125.5, 122.2, 121.6, 118.6, 117.6, 112.1, 45.8, 40.7, 40.5, 40.3, 40.0, 39.8,
39.6, 39.4, 21.0. HRMS (ESI): m/z calcd for C22H21N4O2: 373.1665; found: 373.1673.
Compound 13d: Yield 50%, a pale yellow solid. Purity: 95.4% (by HPLC). MS (ESI) m/z = 373.2
[M+H]+; 1H NMR (d6-DMSO, 400 MHz) δ: 10.96 (s, 1H, NH), 7.94 (d, 1H, J = 8 Hz, Ar-H),
7.61-7.66 (m, 4H, NH, Ar-H), 7.34-7.42 (m, 4H, NH, CH=, Ar-H), 7.04 (m, 2H, Ar-H), 6.58 (d,
1H, J = 16.0 Hz, CH=), 4.88 (m, 2H, CH2), 2.63 (s, 3H, CH3). 13C NMR (d6-DMSO, 100 MHz) δ:
167.0, 151.6, 145.0, 142.2, 141.2, 139.3, 134.4, 131.4, 130.0, 129.4, 129.3, 128.1, 127.8, 127.2,
122.1, 121.6, 118.5, 118.2, 114.4, 112.0, 45.9, 40.7, 40.5, 40.3, 40.0, 39.8, 39.6, 39.4, 21.0. HRMS
(ESI): m/z calcd for C22H21N4O2: 373.1665; found: 373.1677.
Compound 13e: Yield 49%, a pale yellow solid. Purity: 99.1% (by HPLC). MS (ESI) m/z = 449.2
[M+H]+; 1H NMR (d6-DMSO, 400 MHz) δ: 10.81 (s, 1H, NH), 8.04 (d, 1H, J = 8.0 Hz, Ar-H),

ACCEPTED MANUSCRIPT
7.61-7.68 (m, 4H, NH, Ar-H), 7.35-7.55 (m, 7H, NH, CH=, Ar-H), 7.09-7.17 (m, 2H, Ar-H),
6.72-6.75 (m, 1H, Ar-H), 6.58 (d, 1H, J = 16.0 Hz, CH=), 4.61 (m, 2H, CH2), 2.40 (s, 3H, CH3).
13C NMR (d6-DMSO, 100 MHz) δ: 167.2, 151.7, 144.9, 143.3, 143.0, 138.7, 138.0, 136.3, 133.9,
133.0, 129.6, 129.0, 128.7, 128.5, 128.2, 127.7, 122.0, 121.3, 118.9, 117.6, 117.4, 112.5, 45.9,
40.6, 40.4, 40.2, 40.0, 39.8, 39.6, 39.4, 21.4. HRMS (ESI): m/z calcd for C28H25N4O2: 449.1978;
found: 449.1991.
Compound 13f: Yield 53%, a pale yellow solid. Purity: 96.2% (by HPLC). MS (ESI) m/z = 449.2
[M+H]+; 1H NMR (d6-DMSO, 400 MHz) δ: 11.40 (s, 1H, NH), 10.33 (s, 1H, NH), 8.86 (s, 1H,
Ar-H), 8.29 (d, 1H, J = 8.0 Hz, Ar-H), 7.57-7.72 (m, 10H, NH, Ar-H), 7.41-7.47 (m, 2H, CH=,
Ar-H), 6.52 (d, 1H, J = 16.0 Hz, CH=), 4.98 (m, 2H, CH2), 2.39 (s, 3H, CH3). 13C NMR
(d6-DMSO, 100 MHz) δ: 167.0, 151.7, 145.1, 143.0, 141.4, 138.7, 138.0, 136.4, 134.5, 133.8,
130.0, 129.6, 129.5, 129.4, 128.5, 127.8, 127.3, 122.0, 121.3, 118.8, 118.1, 112.5, 46.0, 40.6, 40.4,
40.2, 40.0, 39.8, 39.6, 39.4, 21.4. HRMS (ESI): m/z calcd for C28H25N4O2: 449.1978; found:
449.1969.
Compound 13g: Yield 58%, a pale yellow solid. Purity: 95.3% (by HPLC). MS (ESI) m/z = 465.2
1
[M+H]+; H NMR (d6-DMSO, 400 MHz) δ: 10.94 (s, 1H, NH), 8.03 (d, 1H, J = 8.0 Hz, Ar-H),
7.34-7.68 (m, 11H, NH, CH=, Ar-H), 6.99-7.16 (m, 2H, Ar-H), 6.60 (d, 1H, J = 16.0 Hz, CH=),
4.97 (m, 2H, CH2), 3.70-3.73 (m, 3H, OCH3). 13C NMR (d6-DMSO, 100 MHz) δ: 167.2, 159.8,
152.4, 145.3, 143.0, 140.6, 138.3, 133.7, 132.6, 130.0, 128.8, 128.5, 128.2, 128.0, 121.8, 121.1,
120.9, 117.2, 114.6, 113.6, 112.5, 56.0, 45.8, 40.6, 40.4, 40.2, 40.0, 39.8, 39.6, 39.4. HRMS (ESI):
m/z calcd for C28H25N4O3: 465.1927; found: 465.1941.
Compound 13h: Yield 52%, a pale yellow solid. Purity: 96.6% (by HPLC). MS (ESI) m/z = 465.2
1
[M+H]+; H NMR (d6-DMSO, 400 MHz) δ: 10.82 (s, 1H, NH), 8.05 (d, 1H, J = 8 Hz, Ar-H),
7.34-7.70 (m, 9H, NH, CH=, Ar-H), 7.00-7.19 (m, 3H, Ar-H), 6.71 (m, 1H, Ar-H), 6.58-6.62 (d,
1H, J = 16.0 Hz, CH=), 5.00 (s, 2H, CH2), 3.71 (s, 3H, OCH3). 13C NMR (d6-DMSO, 100 MHz) δ:
167.1, 159.8, 152.4, 143.0, 141.4, 140.6, 138.3, 134.5, 134.2, 133.7, 130.1, 129.3, 128.5, 128.1,
127.8, 127.6, 127.3,127.1, 121.1, 120.9, 119.0, 118.4, 114.5, 114.1, 113.8, 112.5, 55.6, 45.9.
HRMS (ESI): m/z calcd for C28H25N4O3: 465.1927; found: 465.1938.
Compound 13i: Yield 55%, a yellow solid. Purity: 99.4% (by HPLC). MS (ESI) m/z = 465.2
1
[M+H]+; H NMR (d6-DMSO, 400 MHz) δ: 10.73 (s, 1H, NH), 9.01 (s, 1H, Ar-H), 8.39-8.41 (d,
1H, J = 8.0 Hz, Ar-H), 8.15-8.17 (m, 2H, NH, Ar-H), 7.40-7.52 (m, 8H, NH, CH=, Ar-H),
7.18-7.20 (m, 3H, Ar-H), 6.67 (m, 1H, Ar-H), 6.40-6.44 (d, 1H, J = 16.0 Hz, CH=), 4.62 (d, 2H, J
= 4.0 Hz, CH2), 3.89 (s, 3H, OCH3). 13C NMR (d6-DMSO, 100 MHz) δ: 167.2, 152.5, 151.6,
145.2, 143.3, 143.0, 138.6, 133.51, 130.0, 129.7, 128.3, 127.7, 121.2, 114.5, 114.2, 112.5, 55.9,
45.9, 40.6, 40.4, 40.2, 40.0, 39.8, 39.6, 39.4. HRMS (ESI): m/z calcd for C28H25N4O3: 465.1927;
found: 465.1919.
Compound 13j: Yield 50%, a pale yellow solid. Purity: 96.5% (by HPLC). MS (ESI) m/z = 465.2
[M+H]+; 1H NMR (d6-DMSO, 400 MHz) δ: 11.36 (s, 1H, NH), 8.05 (s, 1H, Ar-H), 7.89-7.94 (m,
3H, NH, Ar-H), 7.34-7.67 (m, 7H, NH, CH=, Ar-H), 7.00-7.26 (m, 4H, Ar-H), 6.35-6.39 (d, 1H, J

ACCEPTED MANUSCRIPT
= 16.0 Hz, CH=), 4.96 (s, 2H, CH2), 3.78 (s, 3H, OCH3). 13C NMR (d6-DMSO, 100 MHz) δ:
167.1, 152.5, 151.7, 145.2, 143.0, 141.4, 138.6, 134.4, 134.2, 133.5, 131.8, 130.2, 130.0, 129.7,
128.4, 127.7, 127.2, 121.8, 121.2, 118.9, 118.3, 114.5, 114.2, 112.5, 55.9, 46.0, 40.6, 40.4, 40.2,
40.0, 39.8, 39.6, 39.4. HRMS (ESI): m/z calcd for C28H25N4O3: 465.1927; found: 465.1914.
Compound 13k: Yield 56%, a pale yellow solid. Purity: 95.6% (by HPLC). MS (ESI) m/z = 480.2
[M+H]+; 1H NMR (d6-DMSO, 400 MHz) δ: 11.05 (s, 1H, NH), 8.75 (s, 1H, Ar-H), 8.07-8.39 (m,
5H, NH, Ar-H), 7.61-7.68 (m, 4H, NH, Ar-H), 7.31-7.50 (m, 5H, CH=, Ar-H), 7.13-7.15 (m, 1H,
Ar-H), 6.56-6.60 (d, 1H, J = 16.0 Hz, CH=), 4.98 (m, 2H, CH2). 13C NMR (d6-DMSO, 100 MHz)
δ: 167.2, 152.7, 147.2, 147.1, 145.0, 144.8, 143.4, 143.1, 134.6, 133.0, 130.3, 129.5, 129.0, 128.8,
128.2, 125.8, 124.1, 122.1, 119.3, 117.7, 112.5, 45.8, 40.7, 40.5, 40.3, 40.1, 39.9, 39.6, 39.4.
HRMS (ESI): m/z calcd for C27H21N5O4: 480.1672; found: 480.1680.
Compound 13l: Yield 53%, a pale yellow solid. Purity: 96.3% (by HPLC). MS (ESI) m/z = 480.2
[M+H]+; 1H NMR (d6-DMSO, 400 MHz) δ: 11.14 (s, 1H, NH), 8.74 (s, 1H, Ar-H), 8.27-8.40 (m,
4H, NH, Ar-H), 8.09 (d, 1H, J = 8.0 Hz, Ar-H), 7.35-7.84 (m, 9H, NH, CH=, Ar-H), 7.13 (m, 1H,
Ar-H), 6.51 (d, 1H, J = 16.0 Hz, CH=), 5.02 (s, 2H, CH2). 13C NMR (d6-DMSO, 100 MHz) δ:
167.0, 151.9, 147.2, 145.4, 145.2, 145.0, 143.3, 142.8, 141.2, 135.5, 134.8, 134.5, 130.0, 129.5,
129.0, 128.6, 127.8, 127.4, 125.7, 124.2, 121.2, 119.3, 118.2, 112.5, 45.9, 40.7, 40.5, 40.3, 40.0,
39.9, 39.6, 39.4. HRMS (ESI): m/z calcd for C27H21N5O4: 480.1672; found: 480.1688.
Compound 13m: Yield 51%, a yellow solid. Purity: 95.5% (by HPLC). MS (ESI) m/z = 495.3
1
[M+H]+; H NMR (d6-DMSO, 400 MHz) δ: 10.93 (s, 1H, NH), 8.03 (d, 1H, J = 8 Hz, Ar-H),
7.56-7.69 (m, 4H, NH, Ar-H), 7.30-7.56 (m, 7H, NH, CH=, Ar-H), 7.09-7.11 (m, 2H, Ar-H),
6.58-6.62 (d, 1H, J = 16.0 Hz, CH=), 4.98 (m, 2H, CH2), 3.66-3.83 (m, 6H, 2 × OCH3). 13C NMR
(d6-DMSO, 100 MHz) δ: 167.2, 151.6, 149.6, 149.3, 144.8, 143.4, 143.0, 138.7, 133.8, 132.9,
131.9, 130.3, 128.9, 128.4, 128.0, 127.6, 125.8, 121.9, 121.4, 121.0, 118.8, 117.6, 112.6, 112.2,
112.1, 56.2, 55.7, 42.9, 40.7, 40.5, 40.3, 40.0, 39.9, 39.7, 39.4. HRMS (ESI): m/z calcd for
C29H27N4O4:495.3032; found: 495.3045.
Compound 13n: Yield 47%, a pale yellow solid. Purity: 99.4% (by HPLC). MS (ESI) m/z = 495.3
1
[M+H]+; H NMR (d6-DMSO, 400 MHz) δ: 10.91 (s, 1H, NH), 8.02 (d, 1H, J = 8 Hz, Ar-H),
7.56-7.69 (m, 5H, NH, Ar-H), 7.39-7.54 (m, 5H, NH, CH=, Ar-H), 7.29-7.31 (m, 1H, Ar-H),
7.07-7.11 (m, 2H, Ar-H), 6.57-6.61 (d, 1H, J = 16.0 Hz, CH=), 4.98 (s, 2H, CH2), 3.66-3.83 (m,
6H, 2 × OCH3). 13C NMR (d6-DMSO, 100 MHz) δ: 167.0, 151.7, 149.6, 149.3, 145.0, 143.0,
141.5, 138.7, 134.5, 133.8, 131.9, 129.8, 129.4, 128.4, 127.6, 127.2, 125.8, 122.0, 121.4, 121.0,
118.8, 118.2, 112.6, 112.2, 56.2, 55.7, 42.8, 40.7, 40.5, 40.3, 40.0, 39.9, 39.7, 39.4. HRMS (ESI):
m/z calcd for C29H27N4O4: 495.3032; found: 495.3049.
Compound 13o: Yield 54%, a pale yellow solid. Purity: 98.1% (by HPLC). MS (ESI) m/z = 525.2
[M+H]+; 1H NMR (d6-DMSO, 400 MHz) δ: 10.93 (s, 1H, NH), 8.04 (d, 1H, J = 8.0 Hz, Ar-H),
7.61-7.68 (m, 4H, NH, Ar-H), 7.34-7.49 (m, 4H, NH, CH=, Ar-H), 7.09-7.15 (m, 3H, Ar-H), 6.59
(d, 1H, J = 16.0 Hz, CH=), 4.98 (s, 2H, CH2), 3.71-3.76 (m, 9H, 3 × OCH3). 13C NMR (d6-DMSO,
100 MHz) δ: 167.2, 159.7, 153.4, 144.9, 143.5, 141.5, 138.8, 134.6, 132.9, 130.8, 129.2, 128.5,

ACCEPTED MANUSCRIPT
127.8, 127.3, 121.8, 121.2, 118.9, 117.6, 112.5, 106.0, 56.2, 55.7, 45.9, 40.7, 40.5, 40.3, 40.0, 39.9,
39.7, 39.4. HRMS (ESI): m/z calcd for C30H29N4O5: 525.2183; found: 525.2191.
Compound 13p: Yield 47%, a pale yellow solid. Purity: 95.9% (by HPLC). MS (ESI) m/z = 525.2
[M+H]+; 1H NMR (d6-DMSO, 400 MHz) δ: 10.95 (s, 1H, NH), 8.05 (d, 1H, J = 8.0 Hz, Ar-H),
7.57-7.69 (m, 4H, NH, Ar-H), 7.36-7.47 (m, 5H, NH, CH=, Ar-H), 7.13 (m, 2H, Ar-H), 6.52 (d,
13
1H, J = 16.0 Hz, CH=), 5.00 (s, 2H, CH2), 3.70-3.77 (m, 9H, 3 × OCH3). C NMR (d6-DMSO,
100 MHz) δ: 167.1, 159.7, 152.5, 145.2, 143.0, 141.4, 138.6, 134.2, 133.5, 131.8, 130.2, 129.8,
129.5, 129.2, 128.4, 127.8, 127.2, 121.9, 121.2, 118.8, 118.2, 118.0, 114.4, 112.5, 56.2, 55.7, 46.0,
40.6, 40.4, 40.2, 40.0, 39.8, 39.6, 39.4. HRMS (ESI): m/z calcd for C30H29N4O5: 525.2183; found:
525.2195.
5.4. MTT assay
Human cancer cell lines HepG2, SMMC-7721, HCT116, H1299, Bel7402, and drug-resistant
Bel7402/5-FU were purchased from Shanghai Institute of Cell Biology (China). The in vitro
antiproliferative effects were assessed on related cancer cell lines by the MTT method according
to previous reports [46]. The producing MTT-formazan crystals were melt in DMSO (150 µL).
Then the plates were read through an ELISA plate reader at 570 nm. The inhibitory effect was
expressed as percentage. Corresponding IC50 values were then calculated through GraphPadPrism
(4.03).
5.5. HDAC assay.
According to the instructions of the HDAC fluorimetric assay kit (Enzo Life Sciences Inc.),
the test compounds at different concentrations were respectively incubated with each HDAC
enzyme (HDAC1/3/6/8) in the presence of HDAC substrate (Boc-Lys (Ac)-AMC) at 37 °C for 60
min. The reaction was stopped with the addition of the lysine developer. After 30 min, the
fluorimetric emission was recorded in a fluorescence plate reader with excitation (355 nm) and
emission (460 nm). The HDAC activity was the percentage of activity relative to the control group.
The IC50 was calculated using GraphPad Prism (4.03).
5.6. Cell apoptosis analysis by flow cytometry assay
After overnight culture of Bel7402/5-FU cells, 13p (2.0 and 5.0 µM), SAHA (5.0 µM),
harmine (20 µM), or vehicle were incubated with the cells for 72h. The cells were harvested and
then stained with APC-Annexin V and 7-AAD at 37 °C for 15 min. Flow cytometry analysis
(Calibur, BD, USA) was used to determine the percentage of apoptotic cells, using APC signal
(FL4) (excitation = 633 nm; emission = 660 nm) and 7-AAD staining signal (FL3) (excitation =
488 nm; emission = 647 nm). The data were analyzed using WinList 3D (7.1) and the histogram
was plotted using Excel 2016.
5.7. Acridine orange staining
AO was always used to detect them acidic cellular compartment, which emits bright red
fluorescence in acidic autophagic vacuole vesicles, but green fluoresces in the cytoplasm and
nucleus. After overnight culture of Bel7402/5-FU cells, different concentrations of 13p (2.0 and
5.0 µM), SAHA (5.0 µM), harmine (20 µM), or vehicle were treated with Bel7402/5-FU cells for

ACCEPTED MANUSCRIPT
72h. After treatment, cells were incubated with 1 mg/mL AO for 15 min. Then, the AO was
removed and observed using a fluorescence microscopy.
5.8. Western blotting assay
Western blotting assay was employed to determine the activity of HDAC, apoptosis-related
proteins, and autophagy proteins. Briefly, Bel7402/5-FU cells at 1.5×105/mL were plated in 6 cm
dishes to incubate for one day. Then the cells were treated with or without SAHA, harmine, or 13p
at different concentrations for 72h. After harvested and lyzed, the cell lysates (50 µg/lane) were
separated by SDS-PAGE (12% gel) and transferred onto nitrocellulose membranes. When 5%
fat-free milk was used to block reaction, the target proteins were probed with anti-Ac-histone H3,
anti-Ac-histone H4, anti-Ac-α-tubulin, anti-Bax, anti-Bcl2, anti-H2AX (S139ph), anti-LC3,
anti-p62, and anti-β-actin antibodies (Cell Signaling Technology, MA, USA), respectively. The
bound antibodies were investigated using peroxidase-conjugated secondary antibodies for 1 h, and
then conceived by improved chemiluminescent reagent.
CONFLICT OF INTEREST
The authors declare no conflict of interest, financial or otherwise.
ACKNOWLEDGEMENTS
This work was financially supported by the Natural Science Foundation of China (Nos.
81302628 and 81473089), the Project of “Jiangsu Six Peaks of Talent” (2014-SWYY-044 and
2016-SWYY-CXTD-008), China and Jiangsu Province Postdoctoral Science Foundation
(2018T110533, 2016M590488, and 1601136B), the Project of “Jiangsu 333 high-level talents”,
and also thank Applied Research Projects of Nantong City.
References
[1] World Health Organization Report, 2016. http://www.who.int/cancer/en/.
[2] J.G. Chen, T.W. Kensler, Changing rates for liver and lung cancers in Qidong, China, Chem.
Res. Toxicol. 27 (2014) 3-6.
[3] X. Ma, M. Hu, H. Wang, J. Li, Discovery of traditional Chinese medicine monomers and
their synthetic intermediates, analogs or derivatives for battling P-gp-mediated multi-drug
resistance. Eur. J. Med. Chem. 159 (2018) 381-392.
[4] S. Vijayaraghavalu, C. Peetla, S. Lu, V. Labhasetwar, Epigenetic modulation of the
biophysical properties of drug-resistant cell lipids to restore drug transport and endocytic
functions, Mol. Pharm. 9 (2012) 2730.
[5] M.L. Han, Y.F. Zhao, C.H. Tan, Y.J. Xiong, W.J. Wang, F. Wu, Y. Fei, L. Wang, Z.Q. Liang,
Cathepsin L upregulation-induced EMT phenotype is associated with the acquisition of
cisplatin or paclitaxel resistance in A549 cells, Acta Pharmacol. Sin. 37 (2016) 1606-1622.
[6] J. Huang, H. Mei, Z. Tang, J. Li, X. Zhang, Y. Lu, F. Huang, Q. Jin, Z. Wang,
Triple-amiRNA VEGFRs inhibition in pancreatic cancer improves the efficacy of
chemotherapy through EMT regulation, J. Control Release. 245 (2017) 1-14.

ACCEPTED MANUSCRIPT
[7] (a) H. Kakeya, Natural products-prompted chemical biology: phenotypic screening and a new
platform for target identification, Nat. Prod. Rep. 33 (2016) 648. (b) D.J. Newman, G.M.
Cragg, Natural Products as Sources of New Drugs from 1981 to 2014, J. Nat. Prod. 79 (2016)
629−661.
[8] G.M. Gragg, P.G. Grothaus, D.J. Newman, Impact of natural products on developing new
anti-cancer agents, Chem. Rev. 109 (2009) 3012−3043.
[9] H. Huang, Y. Yao, Z. He, T. Yang, J. Ma, X. Tian, Y. Li, C. Huang, X. Chen, W. Li, S. Zhang,
C. Zhang, J. Ju, Antimalarial β-carboline and indolactam alkaloids from Marinactinospora
thermotolerans, a deep sea isolate, J. Nat. Prod. 74 (2011) 2122.
[10] A. Laine, C. Lood, A. Koskinen, Pharmacological importance of optically active
tetrahydro-β-carbolines and synthetic approaches to create the C1 stereocenter, Molecules. 19
(2014) 1544-67.
[11] R. Cao, W. Peng, Z. Wang, A. Xu, Beta-Carboline alkaloids: biochemical and
pharmacological functions, Curr. Med. Chem. 14 (2007) 479-500.
[12] V. Vishwakarma, J. New, D. Kumar, V. Snyder, L. Arnold, E. Nissen, Q. Hu, N. Cheng, D.
Miller, A.R. Thomas, Y. Shnayder, K. Kakarala, T.T. Tsue, D.A. Girod, S.M. Thomas,
Potent Antitumor Effects of a Combination of Three Nutraceutical Compounds, Sci. Rep. 8
(2018) 12163.
[13] S. Kumar, A. Singh, K. Kumar, V. Kumar, Recent insights into synthetic β-carbolines with
anti-cancer activities, Eur. J. Med. Chem. 142 (2017) 48-73.
[14] Y. Ling, C.J. Xu, L. Luo, J. Cao, J. Feng, Y. Xue, Q. Zhu, C. Ju, F. Li, Y.A. Zhang, X. Ling,
Novel β Carboline/Hydroxamic Acid Hybrids Targeting Both Histone Deacetylase and
DNA Display High Anticancer Activity via Regulation of the p53 Signaling Pathway, J. Med.
Chem. 58 (2015) 9214-9227.
[15] Y. Ling, J. Feng, L. Luo, J. Guo, Y.F. Peng, T.T. Wang, X. Ge, Q. Xu, X. Wang, H. Dai, Y.A.
Zhang, Design and synthesis of novel C3-sustituted β-carboline-based HDAC inhibitors with
potent antitumor activities, Chem. Med. Chem. 12 (2017) 646-651.
[16] Y. Ma, M. Wink, The beta-carboline alkaloid harmine inhibits BCRP and can reverse
resistance to the anticancer drugs mitoxantrone and camptothecin in breast cancer cells,
Phytother Res. 24 (2010) 146-9.
[17] R.G. Fu, Y. Sun, W.B. Sheng, D.F. Liao, Designing multi-targeted agents: An emerging
anticancer drug discovery paradigm, Eur. J. Med. Chem. 136 (2017) 195-211.
[18] P.G. Richardson, P. Moreau, J.P. Laubach, M.E. Maglio, S. Lonial, J. San-Miguel,
Deacetylase inhibitors as a novel modality in the treatment of multiple myeloma, Pharmacol
Res. 117 (2017) 185-191.
[19] H.T. Qin, H.Q. Li, F.Liu, Selective histone deacetylase small molecule inhibitors: recent
progress and perspectives, Expert Opin. Ther. Pat. 27 (2017) 621-636.
[20] K.K. To, W.S. Tong, L.W. Fu, Reversal of platinum drug resistance by the histone deacetylase
inhibitor belinostat, Lung Cancer. 103 (2017) 58-65.
[21] A. Suraweera, K.J. O'Byrne, D.J. Richard, Combination therapy with histone deacetylase
inhibitors (HDACi) for the treatment of cancer: achieving the full therapeutic potential of
HDACi, Front Oncol. 8 (2018) 92.
[22] N. Garmpis, C. Damaskos, A. Garmpi, D. Dimitroulis, E. Spartalis, G.A. Margonis, D.
Schizas, I. Deskou, C. Doula, E. Magkouti, N. Andreatos, E.A. Antoniou, A. Nonni, K.

ACCEPTED MANUSCRIPT
Kontzoglou, D. Mantas, Targeting histone deacetylases in malignant melanoma: A future
therapeutic agent or just great expectations? Anticancer Res. 37 (2017) 5355-5362.
[23] G. Mariño, M. Niso-Santano, E.H. Baehrecke, G. Kroemer. Self-consumption: the interplay
of autophagy and apoptosis, Nat. Rev. Mol. Cell Biol. 15 (2014) 81–94.
[24] M.S. Gilardini Montani, M. Granato, C. Santoni, P. Del Porto, N. Merendino, G. D'Orazi, A.
Faggioni, M. Cirone, Histone deacetylase inhibitors VPA and TSA induce apoptosis and
autophagy in pancreatic cancer cells, Cell Oncol. (Dordr). 40 (2017) 167-180.
[25] H. Rikiishi, Autophagic and apoptotic effects of HDAC inhibitors on cancer cells. J. Biomed.
Biotechnol. 2011 (2011) 830260.
[26] N. Gammoh, D. Lam, C. Puente, I. Ganley, P.A. Marks, X. Jiang, Role of autophagy in
histone deacetylase inhibitor-induced apoptotic and nonapoptotic cell death, Proc. Natl. Acad.
Sci. USA 109 (2012) 6561–6565.
[27] X.Y. Sun, Y. Qu, A.R. Ni, G.X. Wang, W.B. Huang, Z.P. Chen, Z.F. Lv, S. Zhang, H.
Lindsay, S. Zhao, X.N. Li, B.H. Feng, Novel histone deacetylase inhibitor N25 exerts
anti-tumor effects and induces autophagy in human glioma cells by inhibiting HDAC3.
Oncotarget 8 (2017) 75232-75242.
[28] N. Zou, Y. Wei, F. Li, Y. Yang, X. Cheng, C. Wang, The inhibitory effects of compound
Muniziqi granule against B16 cells and harmine induced autophagy and apoptosis by
inhibiting Akt/mTOR pathway. BMC Complement Altern. Med., 17 (2017) 517.
[29] C. Li, Y. Wang, C. Wang, X. Yi, M. Li, X. He, Anticancer activities of harmine by inducing
a pro-death autophagy and apoptosis in human gastric cancer cells. Phytomedicine 28 (2017)
10-18.
[30] T.R. Wilson, P.G. Johnston, D.B. Longley, Anti-apoptotic mechanisms of drug resistance in
cancer, Curr. Cancer Drug Targets. 9 (2009) 307−319.
[31] Y. Gou, Z. Zhang, D. Li, L. Zhao, M. Cai, Z. Sun, Y. Li, Y. Zhang, H. Khan, H. Sun, T.
Wang, H. Liang, F. Yang, HSA-based multi-target combination therapy: regulating drugs'
release from HSA and overcoming single drug resistance in a breast cancer model, Drug
Deliv. 25 (2018) 321-329.
[32] H.Y. Yang, L. Zhao, Z. Yang, Q. Zhao, L. Qiang, J. Ha, Z.Y. Li, Q.D. You, Q.L. Guo,
Oroxylin A reverses multi-drug resistance of human hepatoma BEL7402/5-FU cells via
downregulation of P-glycoprotein expression by inhibiting NF-κB signaling pathway, Mol.
Carcinog. 51 (2012):185-95.
[33] J.H. Lee, J.H. Park, Y. Jung, J.H. Kim, H.S. Jong, T.Y. Kim, Y.J. Bang, Histone deacetylase
inhibitor enhances 5-fluorouracil cytotoxicity by down-regulating thymidylate synthase in
human cancer cells, Mol. Cancer Ther. 5 (2006) 3085-3095.
[34] K. Okada, S. Hakata, J. Terashima, T. Gamou, W. Habano, S. Ozawa, Combination of the
histone deacetylase inhibitor depsipeptide and 5-fluorouracil upregulates major
histocompatibility complex class II and p21 genes and activates caspase-3/7 in human colon
cancer HCT-116 cells, Oncol. Rep. 36 (2016) 1875–1885.
[35] T. Minegaki, A. Suzuki, M. Mori, S. Tsuji, S. Yamamoto, A. Watanabe, T. Tsuzuki, T.
Tsunoda, A. Yamamoto, M. Tsujimoto, K. Nishiguchi, Histone deacetylase inhibitors
sensitize 5-fluorouracil-resistant MDA-MB-468 breast cancer cells to 5-fluorouracil, Oncol.
Lett. 16 (2018) 6202–6208.
[36] R. Hamam, D. Ali, R. Vishnubalaji, Z.F. Alsaaran, E.P. Chalisserry, M. Alfayez, A.

ACCEPTED MANUSCRIPT
Aldahmash, N.M. Alajez, Enhanced efficacy of 5-fluorouracil in combination with a dual
histone deacetylase and phosphatidylinositide 3-kinase inhibitor (CUDC-907) in colorectal
cancer cells, Saudi. J. Gastroenterol. 23 (2017) 34–38.
[37] O. Witt, H.E. Deubzer, T. Milde, I. Oehme, HDAC family: What are the cancer relevant
targets? Cancer Lett. 277 (2009) 8-21.
[38] M. Paris, M. Porcelloni, M. Binaschi, et al., Histone deacetylase inhibitors: From bench to
clinic, J. Med. Chem. 51 (2008) 1505−1529.
[39] J. Chen, Z. Sang, Y. Jiang, C. Yang, L. He, Design, synthesis, and biological evaluation of
quinazoline derivatives as dual HDAC1 and HDAC6 inhibitors for the treatment of cancer.
Chem Biol Drug Des. 2018, 10.1111/cbdd.13405.
[40] S.S. Chauhan, A.K. Singh, S. Meena, M. Lohani, A. Singh, R.K. Arya, S.H. Cheruvu, J.
Sarkar, J.R. Gayen, D. Datta, P.M. Chauhan, Synthesis of novel β-carboline based chalcones
with high cytotoxic activity against breast cancer cells, Bioorg. Med. Chem. Lett. 24 (2014)
2820-2824.
[41] E. White, The role for autophagy in cancer, J. Clin. Invest. 125 (2015) 42-46.
[42] E. White, Deconvoluting the context-dependent role for autophagy in cancer, Nat. Rev.
Cancer. 12 (2012) 401–410.
[43] Y. Lu, R. Zhang, S. Liu, Y. Zhao, J. Gao, L. Zhu, ZT-25, a new vacuolar H(+)-ATPase
inhibitor, induces apoptosis and protective autophagy through ROS generation in HepG2
cells, Eur. J. Pharmacol. 771 (2016) 130-8.
[44] F. Wang, H. Ma, Z. Liu, W. Huang, X. Xu, X. Zhang, α-Mangostin inhibits
DMBA/TPA-induced skin cancer through inhibiting inflammation and promoting autophagy
and apoptosis by regulating PI3K/Akt/mTOR signaling pathway in mice, Biomed.
Pharmacother. 92 (2017) 672-680.
[45] W.S. Laitonjam, B. Thingom, S.D. Moirangthem, Indium-mediated michael addition of
methyl (Z)-2-(Bromomethyl)-3-phenylacrylate to arylidenemalononitriles in aqueous
medium, Org. Prep. Proced. Int. 45 (2013) 246-251.
[46] Y. Ling, J. Guo, Q. Yang, P. Zhu, J. Miao, W. Gao, Y. Peng, J. Yang, K. Xu, B. Xiong, G.
Liu, J. Tao, L. Luo, Q. Zhu, Y. Zhang, Development of novel β-carboline-based hydroxamate
derivatives as HDAC inhibitors with antiproliferative and antimetastatic activities in human
cancer cells, Eur. J. Med. Chem. 144 (2018) 398-409.

ACCEPTED MANUSCRIPT
β-Carboline and N-hydroxycinnamamide hybrids as anticancer
agents for drug-resistant hepatocellular carcinoma
Yong Ling a,b, Wei-Jie Gao a, Changchun Linga,b, Ji Liu a, Chi Meng a, Jianqiang Qiana, Siqun Liua,
Huiling Gan a, Hongmei Wu a, Jinhua Tao a, Hong Dai a,c,*, Yanan Zhang a,*
aSchool of Pharmacy and Jiangsu Province Key Laboratory for Inflammation and Molecular Drug Target, Nantong
University, Nantong 226001, PR China.
bThe Affiliated Hospital of Nantong University, Nantong University, Nantong 226001, PR China.
cCollege of Chemistry and Chemical Engineering, Nantong University, Nantong 226019, PR China
Highlights
1. Novel β-carboline/N-hydroxycinnamamide hybrids 13a-p were designed and synthesized.
2. 13p demonstrated the highest anticancer potency against drug-sensitive and drug-resistant
cells.
3. 13p displayed HDAC1/6 selective inhibition and increased the Ac-H3, Ac-H4 and
Ac-α-tubulin.
4. 13p triggered more potent apoptosis than SAHA by regulating apoptotic proteins expression.
5. 13p induced significant autophagy by up-regulation of LC3-II and down-regulation of LC3-I
and p62.