Hybrids of arenobufagin and benzoisoselenazol reducing the cardiotoxicity of arenobufagin

Article abstract of DOI:10.1016/j.bmcl.2018.08.038

Arenobufagin is a naturally occurring bufadienolide showing promising antitumor activity accompanied however with apparent cardiac toxicity. Following the recent discovery that oxidative damage possibly be an important cause of the cardiac toxicity of cardenolides, a strategy fusing the antitumor agent arenobufagin with a benzoisoselenazol fragment, a reactive oxygen species (ROS) scavenger, has been developed. Six novel hybrids were synthesized and their ROS scavenging activities as well as their in vitro cytotoxicity against the human hepatocellular carcinoma cell line HepG2, an adriamycin-resistant subline HepG2/ADM, and the human myocardial cell line AC16 were evaluated. The results indicate that the hybrids exhibit various degrees of in vitro ROS scavenging activities, and weaker cytotoxicity than that of arenobufagin against the myocardial cell line AC16. These findings suggest the feasibility of a strategy in which the cardiotoxicity of the potential antitumor agent arenobufagin is reduced.

Full text of DOI:10.1016/j.bmcl.2018.08.038

Bioorganic & Medicinal Chemistry Letters xxx (xxxx) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Hybrids of arenobufagin and benzoisoselenazol reducing the cardiotoxicity
of arenobufagin
⁎
Wen Hou, Zhi-Xing Huang, Hong-Gui Xu, Jing Lin , Dong-Mei Zhang, Qun-Long Peng, Hui Lin,
⁎
Yi-Qun Chang, Long-Hai Wang, Zhe Yao, Ping-Hua Sun, Wei-Min Chen
College of Pharmacy, Jinan University, Guangzhou 510632, PR China
A R T I C L E I N F O
A B S T R A C T
Keywords:
Arenobufagin
Benzoisoselenazol
Antitumor
Reactive oxygen species
Hybridization
Arenobufagin is a naturally occurring bufadienolide showing promising antitumor activity accompanied how-
ever with apparent cardiac toxicity. Following the recent discovery that oxidative damage possibly be an im-
portant cause of the cardiac toxicity of cardenolides, a strategy fusing the antitumor agent arenobufagin with a
benzoisoselenazol fragment, a reactive oxygen species (ROS) scavenger, has been developed. Six novel hybrids
were synthesized and their ROS scavenging activities as well as their in vitro cytotoxicity against the human
hepatocellular carcinoma cell line HepG2, an adriamycin-resistant subline HepG2/ADM, and the human myo-
cardial cell line AC16 were evaluated. The results indicate that the hybrids exhibit various degrees of in vitro ROS
scavenging activities, and weaker cytotoxicity than that of arenobufagin against the myocardial cell line AC16.
These findings suggest the feasibility of a strategy in which the cardiotoxicity of the potential antitumor agent
arenobufagin is reduced.
Bufadienolides are a type of natural cardiac steroid with potent
antitumor activity, originally isolated from the Traditional Chinese
Medicine Chan’Su. In our previous studies, we established that com-
pared with other reported bufadienolides isolated from Chan’Su,1
arenobufagin (Fig. 1) possesses the most potent antitumor activity. It
inhibits proliferation of more than ten types of cancer cell lines, espe-
cially the human hepatocellular carcinoma cell line HepG2, with IC50
introduction of ROS scavengers would be a feasible strategy with which
to decrease the cardiac toxicity caused by cardenolides. Therapy with a
combination of ouabain and ROS scavengers in the ouabain-induced
–4
cardiotoxicity models made from isolated heart of guinea pigs were
carried out by the Luo group.1
5–17
Both studies confirmed that the
significant involvement of ROS in ventricular fibrillation caused by
cardenolides, could be ameliorated by ROS scavengers, such as super-
−
10
−8
5
16,17
values ranging from 10
to 10 M. Unfortunately, similar to other
oxide dismutase (SOD) and sodium selenate.
bufadienolides or cardenolides, arenobufagin exhibits strong side ef-
Arenobufagin and other bufadienolides, which are similar to the
cardenolides ouabain and digitoxin, can also bind to Na+/K -ATPase
+
fects, especially cardiac toxicity. Our earlier in vivo toxicology testing
1
on mice showed its LD50 to be as low as 20 mg/kg. and at least one
and display similar structure-activity relationships (SAR) regarding
6
18–21
lethal case has been reported following ingestion of Chan’Su tea. Over
their antitumor activity.
In view of these findings, we planned to
decades, numerous researchers have developed bufadienolide analo-
gues by chemical or biological transformation. Most of these efforts
focused on enhancement of antitumor activity and put less emphasis on
reducing the cardiac toxicity, which significantly affects the safety of
combine arenobufagin with an ROS scavenger fragment, in an effort to
obtain arenobufagin derivatives with improved clinical efficacy and
safety.
Benzoisoselenazol (BISA) derivatives represented by Ebselen
(Fig. 2), is by far one of the best mimics of GSH peroxidase. Numerous
in vitro and in vivo experiments have established that BISA derivatives
2
,7–12
their clinical use.
Consequently, identification of the mechanisms
13
underlying the cardiac toxicity and discovery of methods to reduce it
can affect the future of arenobufagin or other bufadienolide derivatives.
Recently, Ho et al. employed a combination of pharmacological
tools and genetic mouse models to investigate the mechanisms of
generation of reactive oxygen species (ROS), which are involved in the
adverse pro-arrhythmic effects of digitoxin.1 Their study revealed
protect effectively against oxidative challenges including those from
ROS.2
2,23
Amino acids are known to be capable of promoting pharma-
24–26
cological bioavailability in vivo,
and the effect of combining BISA
fragments with free amino acids or their esters has been examined by
4
27–30
many research groups exploring efficient ROS scavengers.
Natural
⁎
Corresponding authors.
E-mail addresses: linjing_jnu@163.com (J. Lin), twmchen@jnu.edu.cn (W.-M. Chen).
https://doi.org/10.1016/j.bmcl.2018.08.038
Received 8 May 2018; Received in revised form 26 August 2018; Accepted 27 August 2018
0960-894X/ © 2018 Published by Elsevier Ltd.
Please cite this article as: Hou, W., Bioorganic & Medicinal Chemistry Letters (2018), https://doi.org/10.1016/j.bmcl.2018.08.038

W. Hou et al.
Bioorganic & Medicinal Chemistry Letters xxx (xxxx) xxx–xxx
HepG2 of 7.301. Therefore, we chose phenylalanine as the linker in
another three compounds (6d–6f) designed to evaluate the cytotoxic
influences of substituents at C-4 of the BISA aromatic ring. This series
contains three substituents from electron-donating groups
(
6d > 6c > 6e) and one from an electron-withdrawing group (6f).
The synthesis route to these compounds includes six steps shown in
Scheme 1 and allow for the synthesis of all six hybrids whose structures
were established by 1H and C NMR spectroscopy as well as low- and
high-resolution electrospray ionization (ESI) mass spectrometry (Detail
data were in the Supporting information).
13
Fig. 1. Chemical structures of cardenolides, bufadienolides and arenobufagin.
Supplementary data associated with this article can be found, in the
online version, at https://doi.org/10.1016/j.bmcl.2018.08.038.
As shown in Scheme 1, 2-Aminobenzoic acid was used as a start ma-
terial to obtain 2a–2d. Then 2a–2d reacted with thionyl chloride to offer
amino acid derivatives could be employed as a linker at C-3 of areno-
bufagin to connect arenobufagin and a BISA fragment, and according to
our previous SAR study of bufadienolides.31 a small decrease in the in
3a–3d, which were used for the next step quickly for their instability. N-
vitro cytotoxicity can result.
((benzyloxy)carbonyl)amino acid was added to thionyl chloride to give
In the present study, novel hybrids of BISA fragment and arenobu-
fagin were designed, synthesized and subjected to preliminary assess-
ment of their cytotoxicity, as well as their ability to scavenge ROS.
In order to reveal the influence of various amino acids on the in vitro
cytotoxicity, three typical linkers including an aromatic amino acid,
phenylalanine, and aliphatic amino acids, glycine and isoleucine, were
employed in the design of hybrids (6a–6c). We found that phenylala-
nine as the linker (6c) is the most beneficial, with a selectivity index for
carbonyl chloride derivatives which was esterified by arenobufagin
quickly. Then the protecting group on the esterification products (4a–4c)
were deprotected to provide 5a–5c. Finally, 3a–3d were reacted with
5a–5c to obtain the finally products 6a–6f. The antitumor activities of the
target compounds were first evaluated. HepG2 and HepG2/ADM were
used in an MTT (3-(4,5-dimethyl-2-thiazolyl)-2,5-diphenyl-2H-tetrazolium
bromide) assay. Consistent with our previous SAR study, an acceptable
Fig. 2. Chemical structures of benzoisoselenazol derivatives, ebselen and amino acid derivatives of BISA.
Scheme 1. General synthetic route to arenobufagin derivatives 1–6. Reagents and conditions: (i) NaNO
HOBt, DMAP; (v) PdCl , Et SiH; (vi) Et N, CH Cl
2 2 2 2
, HCl; (ii) Na Se ; (iii) SOCl , reflux; (iv) arenobufagin, EDCI,
2
3
3
2
2
.
2

W. Hou et al.
Bioorganic & Medicinal Chemistry Letters xxx (xxxx) xxx–xxx
Table 1
Inhibition of HepG2, HepG2/ADM and AC16 by the target compounds (6a–6f).
Compounds
IC50 (μM) ± SD
Selectivity index for HepG2
HepG2
HepG2/ADM
AC16
6
6
6
6
6
6
a
b
c
d
e
f
0.607 ± 0.020
14.530 ± 0.51
0.489 ± 0.450
0.305 ± 0.006
0.339 ± 0.551
0.265 ± 0.076
0.020 ± 0.001
2.447 ± 0.219
36.713 ± 0.988
7.943 ± 1.840
5.859 ± 0.420
5.589 ± 0.534
9.742 ± 0.735
0.056 ± 0.003
2.130 ± 0.691
14.990 ± 0.820
3.570 ± 1.113
2.200 ± 0.981
3.210 ± 0.815
4.303 ± 0.701
0.005 ± 0.002
3.509
1.032
7.301
7.213
9.469
16.238
0.250
Arenobufagin
All dates represent mean ± S.D. from different experiments performed in triplicate.
downtrend in the bioactivity was observed. As shown in Table 1, all tested
compounds still possess inhibitory effects on HepG2 and HepG2/ADM,
with IC50 values at the submicromolar and micromolar range, respectively.
The type of amino acid linker used has a significant influence on the an-
titumor activity. Compound 6b, which possesses an amino acid with a
large aliphatic side chain, the isobutyl group, was found to have an IC50
value against the HepG2 cell line of 14.530 μM, compared to that of its
corresponding compound 6d (0.305 μM). The above results suggest that a
small steric linker (6a) and an aromatic linker (6c–6f) seem to be bene-
ficial to the in vitro anticancer effect. No remarkable difference of the IC50
values of compounds 6c–6f was observed. It was thought that substitutes
at C-4 of the BISA aromatic ring might not be key in influencing the in vitro
cytotoxicity. Meanwhile, the AC16 cell line was selected to test whether
introduction of ROS scavenger (BISA) can reduce the damage caused by
arenobufagin to myocardial cells. Our results (Table 1) revealed that all
hybrids have low cytotoxicity against the AC16 cell (with IC50 values
ranging from 2.130 to 14.990 μM) compared to that of arenobufagin
Fig. 3. The photographs of AC16 cells after and before 1 μM of compounds were
added for 24 h. (a) DMSO treated group; (b) 1 μM arenobufagin treated group;
(
IC50 = 0.005 μM). Thus the compounds are less toxic to the human AC16
cell than against cancer cell lines. For the HepG2 cell line, they are rela-
tively selective with a selectivity index ranging from 7.213 to 16.238 for
(
c) 1 μM 6c treated group; (d) 1 μM 6d treated group; (e) 1 μM 6e treated group;
f) 1 μM 6f treated group.
(
6c–6f, and 1.032 for 6b, and 3.509 for 6a, which are much better than the
selectivity index of arenobufagin (0.250). These results demonstrate that
our modified compounds are less cytotoxic then arenobufagin itself.
In order to further confirm the toxic effects of these hybrids on
cardiomyocytes, the impact of the hybrids 6c–6f, with a good se-
lectivity index on cell morphology were also examined using the human
myocardial cell line AC16. AC16 cells were treated with 1 μM of hybrids
6c–6f and arenobufagin for one day. Cells were observed and photo-
graphs were taken with a phase contrast microscope. It can be inferred
from Fig. 3, that treatment with the compounds (6c–6f) with good se-
lectivity index fails to result in determinant morphological changes
compared to the control (Fig. 3c–f vs a). However, obvious cell swelling
and detachment from the substratum is observed after the treatment
with 1 μM arenobufagin (Fig. 3b). The results reflect that our com-
pounds are less toxic than arenobufagin to the AC16 cell line.
Therewith, to verify whether our hybrids could scavenge ROS in vitro
as expected, an ROS scavenger assay was conducted for 6c–6f, which
maintain good antitumor activity and show less cardiomyotoxicity and
cytotoxicity. Arenobufagin was used as a negative control and ebselen as a
positive control. As shown in Fig. 4, after the addition of ebselen, a slight
reduction of ROS is observed, while ROS augmentation is detected in the
arenobufagin group. All the selected hybrids (6c–6f) were found to ap-
propriately scavenge ROS in HepG2, and compounds 6d and 6f are sig-
nificantly different from members of the arenobufagin group, indicating
that a combination of arenobufagin and BISA contributes to ROS reduc-
tion. Unfortunately, compound 6c fails to scavenge ROS as expected but
most hybrids can diminish the ROS in HepG2 as a result of the introduc-
tion of the BISA fragment, which may be a key factor in lowering the AC16
cell cytotoxicity of our novel compounds. Therefore we tentatively infer
that our compounds may show some decline in cardiotoxicity in vivo, and
this merits further investigation in the future.
Fig. 4. ROS scavenging ability on HepG2 of the tested compounds. The tested
concentration was 1 μM. 20 μM dichlorofluorescin diacetate (DCFH-DA) in-
cubation lasting 20 min was necessary after 2 h interaction between compounds
and HepG2 cancer cells.
Finally, we have applied H
model to observe the ROS elimination ability of compounds 6c–6f
incubation for 6 h resulted in
2 2
O -induced oxidative damage AC16
2 2
(Fig. 5). Results shown that 100 μM H O
severe damage to AC16 cells (Fig. 5b vs 5a). If AC16 cells were pre-
treated with 1 μM ebselen, 6c–6f for 30 min, cell damage induced by
2 2
100 μM H O was reduced, especially in 1 μM ebselen, 6d, 6f group
3

W. Hou et al.
Bioorganic & Medicinal Chemistry Letters xxx (xxxx) xxx–xxx
Fig. 5. The photographs of AC16 cells or H
ebselen, 6c, 6d, 6e, or 6f for 30 min, then 100 μM H O
2 2
2
O
2
-AC16 cells, (a) DMSO treated group; (b) H
were added for another 6 h.
2
O
2
-AC16 group; (c-h) AC16 cells were pre-treated with 1 μM arenobufagin,
(
1
Fig. 5d–h vs 5b). However, severe cell damage was also observed in
μM arenobufagin group (Fig. 5c vs 5a or 5b). These results demon-
display similar in vitro growth inhibitory effects in cancer cells andbinding profiles to
the alpha subunits of the Na+/K -ATPase. Mol Cancer. 2013;12:33–47.
0. Borges KB, Borges WD, Duran-Patron R, Pupo MT, Bonato PS, Collado IG.
Stereoselectivebiotransformations using fungi as biocatalysts. Tetrahedron
Asymmetry. 2009;20:385–397.
+
1
strated that compounds ebselen, 6d, 6f could obviously reverse the
-induced oxidative damage to AC16 cell, implying their well ROS
2 2
H O
1
1. Tian HY, Luo SL, Liu JS, et al. C-23 steroids from the venom of Bufo bufogargarizans.
J Nat Prod. 2013;76:1842–1847.
eliminate ability.
To the best of our knowledge, this is the first major study of the
design of hybrids of arenobufagin and BISA fragment. Six novel BISA-
containing arenobufagin derivatives employing a natural amino acid
linker (6a–6f) were synthesized. As expected, their in vitro cytotoxicity
against HepG2 and HepG2/ADM were in the submicromolar-range and
micromolar-range respectively, in agreement with our previous in vitro
cytotoxic SAR study. More importantly, the cytotoxicity against human
AC16 cell of our derivatives is weaker than that of arenobufagin.
Meanwhile, we have established that our compounds scavenge ROS in
1
2. Zhu ZT, Deng S, Liu D, et al. Isolation and identification of phase I metabolitesof
resibufogenin in rats. Xenobiotica. 2013;43:479–485.
13. Abdelrahman MA, Ahmed SH, Nabil ZI. In vitro cardiotoxicity and mechanism of
action of the Egyptian green toad Bufo viridis skin secretions. Toxicol In Vitro.
2
010;24:480–485.
14. Ho HT, Liu B, Snyder JS, et al. Ryanodine receptor phosphorylation byoxidized
CaMKII contributes to the cardiotoxic effects of cardiac glycosides. Cardiovasc Res.
2
014;101:165–174.
1
5. Chen ZF, Ji XL, Luo Q, Luo SH. The cardiotoxicity of free radicals and cardiac gly-
cosides. Chin Pharmacol Bull. 1992;8:107–110.
16. Chen ZF, Huang RB, Ji XL, Luo Q. The anti-arrhythmiabrought by uabainstudy of free
radical scavenging agents inguinea pigs. Chin Pharmacol Bull. 1991;7:117–120.
2 2
HepG2 and H O -AC16. In conclusion, a strategy to reduce the cardi-
otoxicity of the potential antitumor agent arenobufagin and several
excellent antitumor candidates has been defined by this study.
1
7. Chen ZF, Ji XL, Luo Q. The study of influence on toxicity to the isolated guinea pig
brought by uabain of some oxygen free radical scavenging agent. J Jinan Univ (Nat
Sci Med Ed). 1991;12:7–11.
1
1
2
2
8. Ma BA, Xiao ZY, Chen YJ, et al. Synthesis and structure-activity relationshipsstudy of
cytotoxic bufalin 3-nitrogen-containing-ester derivatives. Steroids. 2013;78:508–512.
9. Gomes A, Bhattacharjee P, Mishra R, et al. Anticancer potential of animalvenoms and
toxins. Indian J Exp Biol. 2010;48:93–103.
Acknowledgements
0. Cruz Jd.S, Matsuda H. Arenobufagin, a compound in toadvenom, blocks Na+-K+
This work was supported by the Natural Science Foundation of
Guangdong Province (No. S2012010008130), the Research Fund for the
Doctoral Program of Higher Education of China (No. 20124401110008),
and the Natural Science Foundation of China (No. 81402774).
pump current in cardiac myocytes. Eur J Pharmacol. 1993;239:223–226.
1
.
C
r
u
z
J
D
S
,
M
a
t
s
u
d
a
H
.
D
e
p
r
e
s
s
i
v
e
e
ff
e
c
t
s
o
f
a
r
e
n
o
b
u
f
a
g
i
n
o
n
t
h
e
d
e
l
a
y
e
d
r
e
c
t
i
fi
e
r
K
+
current of guinea-pig cardiac myocytes. Eur J Pharmacol Mol Pharmacol.
1994;266:317–325.
2. He J, Li D, Xiong K, et al. Inhibition of thioredoxin reductase by a novel series of bis-
2
1
,2-benzisoselenazol-3(2H)-ones: Organoselenium compounds for cancer therapy.
References
Bioorg Med Chem. 2012;20:3816–3827.
3. Parnham M, Sies H. Ebselen: prospective therapy for cerebral ischaemia. Expert Opin.
Invest. Drugs. 2000;9:607–619.
4. Dragzalesinska M, Kulbacka J, Saczko J, et al. Esters of betulin and betulinic acid
with amino acids have improved water solubility and are selectively cytotoxic to-
ward cancer cells. Bioorg Med Chem Lett. 2009;19:4814–4817.
5. A.M. Landel W.G. Hammond M.M. Meguid Aspects of amino acid and protein me-
tabolism in cancer-bearing states Cancer 55 2015 230 237.
6. Jr SM, Daniels GA, Boerner P, Lin J. Neutral amino acid transport systems in animal
cells: potential targets of oncogene action and regulators of cellular growth. J Membr
Biol. 1988;104:1–20.
7. Xu HS, Liu XF, Xiao YX. Synthesis and anti-lipid peroxidation of the amino acid
derivatives of benzene and isoselenazone. Chin. J. Org. Chem. 1995;15:252–258.
8. Xiao YX, Liu XF, Xu HS, Sun SY, Xu B. Synthesis and anti-lipid peroxy of ebselen
amino acid derivatives. Chem. Res. Chin. Univ. 1996;17:914–916.
9. Zeng HH. Benzoisoselenazole derivatives with anti-inflammation, antivirus and an-
tithrombosis activity and their use. US Pat., US8609702B2; 2013, pp. 1–10.
0. Elsherbini M, Hamama WS, Zoorob HH, Bhowmick D, Mugesh G, Wirth T. Synthesis
and antioxidant activities of novel chiral ebselen analogues. Heteroat Chem.
2
2
1
.
.
Ye WC, Zhang DM, Liu JSh, Cao HH, Tian HY. The Application of Arenobufagin in
Antitumour Agents. PR China Pat CN102038690B; 2012, p. 1–15.
2
Deng LJ, Wang LH, Peng CK, et al. Fibroblast activation protein α activated tripeptide
bufadienolide antitumor prodrug with reduced cardiotoxicity. J Med Chem.
2
2
2
017;60:5320–5333.
3
.
Tian HY, Wang L, Zhang XQ, et al. Two novel 19-norbufadienolides with un-
precedented skeletons from the venom of Bufo bufogargarizans. Chem Eur J.
2
010;16:10989–10993.
4
.
.
Tian HY, Wang L, Zhang XQ, et al. New bufadienolides and C-23 steroids fromthe
venom of Bufo bufo gargarizans. Steroids. 2010;75:884–890.
2
2
2
3
5
Zhang DM, Liu JS, Deng LJ, et al. Arenobufagin, a natural bufadienolide fromtoad
venom, induces apoptosis and autophagy in human hepatocellular carcinomacells
through inhibition of PI3K/Akt/mTOR pathway. Carcinogenesis. 2013;34:1331–1342.
Ko RJ, Greenwald MS, Loscutoff SM, et al. Lethal ingestion of Chinese herbal tea
containing Chan’Su. West J Med. 1996;164:71–75.
Heasley B. Chemical synthesis of the cardiotonic steroid glycosides and related nat-
ural products. Chem Eur J. 2012;18:3092–3120.
Gao HM, Popescu R, Kopp B, Wang ZM. Bufadienolides and their antitumoractivity.
Nat Prod Rep. 2011;28:953–969.
Banuls LMY, Katz A, Miklos W, et al. Hellebrin and its aglycone formhellebrigenin
6.
7.
8.
9.
2
015;25:320–325.
3
1. Xu HG, Yao Z, Lin J, Chen WM. Progress in the structure-activity relationship of
bufadienolides as antitumor agent. Chin Sci Technol Paper Online. 2014.
4