Natural abenquines and synthetic analogues: Preliminary exploration of their cytotoxic activity

Article abstract of DOI:10.1016/j.bmcl.2017.01.079

In this study, we explore the cytotoxic activity of four natural abenquines (2a–d) and fourteen synthetic analogues (2e–j and 3a–h) against a panel of six human cancer cell lines using a SRB assay. It was found that most of the compounds revealed higher levels of cytotoxic activities than naturally occurring abenquines. The analogues carrying ethylpyrrolidinyl and ethylpyrimidinyl with either an acetyl group (2?h–i) or a benzoyl group (3f–g), were the most potent against all human cancer cell lines and displayed EC₅₀ between a range of 0.6–3.4?μM. Notably, of the compounds tested, compound 2i proved the most cytotoxic against both ovarian (A2780) and breast (MCF7) cells, showing EC₅₀?=?0.6 and 0.8?μM respectively. Likewise, the analogues 2i, 3f and 3?g showed strong activity against cell HT29 with EC₅₀?=?0.9?μM for these compounds.

Full text of DOI:10.1016/j.bmcl.2017.01.079

Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Natural abenquines and synthetic analogues: Preliminary exploration
of their cytotoxic activity
a
c
Amalyn Nain-Perez a, Luiz C.A. Barbosa a,b, , Diego Rodríguez-Hernández , Annemarie E. Kramell ,
⇑
Lucie Heller c, René Csuk c,
⇑
a Department of Chemistry, Universidade Federal de Minas Gerais, Av. Pres. Antônio Carlos, 6627, Campus Pampulha, CEP 31270-901, Belo Horizonte, MG, Brazil
b Department of Chemistry, Universidade Federal de Viçosa, Av. P.H. Rolfs, s/n, CEP 36570-900, Viçosa, MG, Brazil
c Martin-Luther-University Halle-Wittenberg, Organic Chemistry, Kurt-Mothes-Str.2, D 06120 Halle (Saale), Germany
a r t i c l e i n f o
a b s t r a c t
Article history:
In this study, we explore the cytotoxic activity of four natural abenquines (2a–d) and fourteen synthetic
analogues (2e–j and 3a–h) against a panel of six human cancer cell lines using a SRB assay. It was found
that most of the compounds revealed higher levels of cytotoxic activities than naturally occurring aben-
quines. The analogues carrying ethylpyrrolidinyl and ethylpyrimidinyl with either an acetyl group (2 h–i)
or a benzoyl group (3f–g), were the most potent against all human cancer cell lines and displayed EC50
Received 18 November 2016
Revised 24 January 2017
Accepted 25 January 2017
Available online xxxx
between a range of 0.6–3.4
toxic against both ovarian (A2780) and breast (MCF7) cells, showing EC50 = 0.6 and 0.8
Likewise, the analogues 2i, 3f and 3 g showed strong activity against cell HT29 with EC50 = 0.9 lM for
these compounds.
l
M. Notably, of the compounds tested, compound 2i proved the most cyto-
M respectively.
Keywords:
Abenquines
Aminoquinone
Abenquines analogues
Cytotoxicity
l
Ó 2017 Elsevier Ltd. All rights reserved.
SRB assay
The global human population is projected to reach nine billion
people by 2050, and this will result in an unprecedented challenge
for everyone.1,2 Together with population growth, the incidence of
certain types of cancer is expected to increase due to risk-factors
that affect human health as obesity, tobacco and alcohol
consumption.3
Amongst the major diseases that afflict humanity, cancer has
become a major cause of global mortality, with approximately
12.7 million new cases diagnosed every year. According to the
World Health Organization, in 2012 approximately 14 million
new cases of morbidity and mortality and 8.2 million cancer
related cases were reported worldwide. As estimated, the number
of new cases could rise to around 70% in the next two decades.4
With the growing demand for treatments of cancer, the devel-
opment of new effective drugs with fewer side effects is a priority
for big pharmaceutical industries.5 Historically, the use of natural
products for the treatment and control of various diseases, includ-
ing cancer is well documented.6 Currently, 34% of commercial
medicines are based on small molecules such as natural products
or their derivatives which represent a broad range of chemical
structures.6,7 Amongst these, naturally occurring quinones are
widely distributed across plants, animals and bacteria.8,9 Among
the quinones, several aminoquinones display significant and
diverse biological properties. For instance, streptonigrin,10,11 gel-
danamycin12,13 and mitomycin C14,15 have shown potent antitu-
mor and/or antibiotic activities. In 1987, the derivative of
mitomycin C known as EO9 was developed as a targeted therapeu-
tic agent, and clinical evaluation revealed significant anti-tumor
activity.16
Recently, a new group of aminoquinones called abenquines A-D
(Fig. 1) were isolated from Streptomyces sp. strain DB634, collected
from the soils of the Chilean highland of the Atacama Desert.17
These compounds possess an N-acetyl-aminobenzoquinone core
bearing natural amino acid residues. The abenquines showed low
cytotoxic activity against selected bacteria and dermatophytic
fungi and were found to be phosphodiesterase type 4 (PDE4b)
inhibitors.17 Recently, these natural compounds were synthesized
by our group and showed phytotoxic activity against the cyanobac-
terium Synechococcus elongates.18
During the last few years, our group has been using natural
products as scaffold for the development of new bioactive sub-
stance with potential application as pharmaceuticals or agrochem-
icals.19–23 Hence, in order to gain a deeper insight into the
biological profile of this poorly investigated group of quinones
we now report the cytotoxic profiles of four natural abenquines
and fourteen synthetic analogues against several cancer lines.
⇑
Corresponding authors at: Department of Chemistry, Universidade Federal de
Minas Gerais, Av. Pres. Antônio Carlos, 6627, Campus Pampulha, CEP 31270-901,
Belo Horizonte, MG, Brazil (L.C.A. Barbosa).
E-mail addresses: lcab@ufmg.br (L.C.A. Barbosa), rene.csuk@chemie.uni-halle.de
(R. Csuk).
http://dx.doi.org/10.1016/j.bmcl.2017.01.079
0960-894X/Ó 2017 Elsevier Ltd. All rights reserved.
Please cite this article in press as: Nain-Perez A., et al. Bioorg. Med. Chem. Lett. (2017), http://dx.doi.org/10.1016/j.bmcl.2017.01.079

2
A. Nain-Perez et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
amides were oxidized using phenyliodine diacetate (PIDA), pro-
viding quinones 2 (with acetyl group) in 75% and 3 (with ben-
zoyl group) in 60% yield. Treatment of quinones 2 and 3 with
different amino acids in the presence of sodium bicarbonate in
ethanol for two hours at reflux, gave the required abenquine
(2a–d) and various analogues (2e–f and 3a–d) in variable yields
(45–81%). This reaction involves a Michael addition of the amino
group at the quinone core, resulting in an aminodihydrobenzo-
quinone intermediate. This intermediate can be oxidized by the
starting benzoquinone if used in excess24,25 or can be sponta-
neously oxidized by air.18
Fig. 1. Structural formulas of Abenquines isolated from Streptomyces sp.
The analogues 2g–j and 3e–h were prepared via a reaction of
equivalent amount of the required amine and quinones 2 or 3 in
dichloromethane, at room temperature, for two hours. The
required compounds were obtained in overall yields ranging from
29% to 64% from 1. Compounds 2a–f have previously been pre-
pared by our group.18
The natural abenquine and analogues were prepared as
shown in Scheme 1, following a recently reported procedure.18
Briefly, commercially available 2,5-dimethoxyaniline 1 was con-
verted into an amide by the reaction of the respective acetyl
or benzoyl chloride in quantitative yield. Subsequently, the
Scheme 1. Structures of natural abenquines (2a–d) and their analogues (2e–j; 3a–h).
Please cite this article in press as: Nain-Perez A., et al. Bioorg. Med. Chem. Lett. (2017), http://dx.doi.org/10.1016/j.bmcl.2017.01.079

A. Nain-Perez et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
3
Table 1
Cytotoxicity for abenquines (2a–d) and analogues (2e–j and 3a–h) (cut-off in all experiments was at 30
fibroblasts (NIH 3T3); betulinic acid (BA) was used as the standard.
lM); employing human tumor cell lines and non-malignant mouse
a
Compound
EC50
Melanoma (518A2)
Ovarian (A2780)
Colon (HT29)
Breast (MCF7)
Lung (A549)
Hypopharyngeal (FaDu)
NIH 3T3
2a
2b
2c
2d
2e
2f
2g
2h
2i
>30
>30
>30
>30
7.1
15.4
9.7
3.4
>30
>30
>30
>30
5.7
5.9
3.5
1.0
0.6
>30
>30
>30
>30
12.7
26.8
18.6
1.4
>30
>30
>30
>30
7.3
15.2
8.6
0.9
>30
>30
>30
>30
6.5
13.7
7.4
2.8
>30
>30
>30
>30
11.7
16.7
15.2
2.2
>30
>30
>30
>30
6.6
7.8
6.6
1.5
1.2
2.5
0.9
0.8
2.0
1.8
2j
9.6
1.4
4.4
2.2
2.6
5.8
6.3
3a
3b
3c
3d
3e
3f
>30
>30
>30
>30
n.s.b
1.7
>30
>30
>30
>30
n.s.
1.4
>30
>30
>30
>30
n.s.
0.9
>30
>30
>30
>30
n.s.
1.5
>30
>30
>30
>30
n.s.
1.5
>30
>30
>30
>30
n.s.
1.7
>30
>30
>30
>30
n.s.
2.0
3g
3h
BA
1.7
6.5
13.6
1.3
5.5
12.7
0.9
3.5
18.4
1.5
6.2
11.4
1.4
5.4
7.6
1.7
7.8
13.7
2.1
7.5
13.1
a
EC50 values in lM from SRB assay after 96 h of treatment; the values are averaged from at least three independent experiments performed each in triplicate; confidence
interval CI = 95% (individual positive (upper values) and negative (lower values) errors are given in the Table S1 of supplementary part.
b
n.s. not soluble – the assay could not be carried out.
The cytotoxic activities of abenquines and their analogues (2a–j
and 3a–h) were evaluated against six human tumor cell lines
(namely: melanoma cells - 518A2; ovarian carcinoma - A2780;
colon adenocarcinoma - HT29; breast adenocarcinoma - MCF7;
lung cancer - A549; hypopharyngeal carcinoma - FaDu) and non-
malignant mouse fibroblasts (NIH 3T3) using the well-established
colorimetric sulforhodamine B assay.26 The results of these assays
are summarized in Table 1.
As observed from Table 1, compound 3e bearing both benzoyl
and benzyl groups proved insoluble under the conditions of the
SRB assay, and could not be assayed. Natural abenquines (2a–d)
and their corresponding benzoyl analogues, (3a–d), showed low
cytotoxicity (EC50 > 30 lM = cut-off of the assay) against the cells
Fig. 2. Cytotoxicity (EC50 values in lM with standard error (SE)), from SBR assay for
abenquines analogues 2h, 2i, 3f and 3g for all lines tested.
lines tested. Interesting, compound 2e (a d-valine derivative) an
isomer of abenquine C (2c) revealed an increase in cytotoxic activ-
ity. These results suggested that analogues of abenquine possess-
ing an amino acid residue with d-configuration could modulate
the cytotoxic activity of these compounds. However, the synthetic
analogue derived of l-methionine (2f) also showed a slightly higher
activity compared to the natural abenquines. These data indicate
that not only the stereochemistry of the amino acids but also its
structure can influence the cytotoxicity.
fibroblasts (NIH 3T3) (Table S1). Similarly, the compound 2i was
the most active against this line cell (A2780) with EC50 = 0.6
but showed not selectivity.
lM,
In general, some abenquine analogues have showed to be more
potent than other aminoquinone natural cytotoxic, such as naki-
jiquinones or mansouramycin and their analogues.27–29
Further analyses of Table 1 revealed that analogues bearing an
extra amino group (compounds 2i, 2h, 3f, 3g) were the most active
compounds, as illustrated by the selected data presented in Fig. 2.
Such compounds have in common the ethylpyrrolidinyl (2h, 3f) or
ethylpyrimidinyl (2i, 3g) group. They differ from the acetyl (2h and
2i) and benzoyl (3f and 3g) groups. Since all these four compounds
showed EC50 values in the same order of magnitude, ranging
To obtain preliminary indications of the mode of action of this
class of compounds, analogues 2i and 3g were subjected to addi-
tional experiments using an acridine orange/propidium iodide
assay (AO/PI).30 Fig. 3 shows the results from the fluorescence
microscopic analysis of ovarian cancer cells (A2780). For com-
pound 2i only, a decrease in cell number was observed after treat-
ment with a concentration of 1.2
lM, while the cells treated with
between 0.6 and 3.4 lM (Fig. 2), we conclude that the nature of
3g (2.6 M) showed a higher level of orange stained cells indicat-
l
the acetyl group is less relevant for cytotoxicity. These results indi-
cated that compounds 2h, 2i, 3f and 3g are at least ꢀ20 times more
active than the natural abenquines (2a–d); however, they were
toxic against non-malignant mouse fibroblasts cells (NIH 3T3)
showing low selectivity (Table S1).
Other analogues, such as 2g, 2j and 3h, showed good cytotoxi-
city (Table 1). For example, an EC50 = 3.5 and 1.4 lM, were found
for compounds 2g and 2j with the ovarian cancer cell line
(A2780) and only the compound 2j showed some selectivity
(SI = 4.5) between this cell line (A2780) and non-malignant mouse
ing a loss of membrane integrity. Measurement of 2i using a 4-
times EC50 concentration indicated also the presence of late-stage
apoptotic cells. Nevertheless, it is widely known that the amino-
quinone compounds has two important properties: firstly, the qui-
nonic moiety can be reduced through one or two electron and;
secondly, this has an alkylating (amino) group that can form cova-
lent bonds with a variety of cellular components, the oxidation
state of quinone being able to modulate the activity of the alkylat-
ing element.8 Thus, the cytotoxic activity shown by some aben-
quine analogues could be attributed to redox cycling.31
Please cite this article in press as: Nain-Perez A., et al. Bioorg. Med. Chem. Lett. (2017), http://dx.doi.org/10.1016/j.bmcl.2017.01.079

4
A. Nain-Perez et al. / Bioorganic & Medicinal Chemistry Letters xxx (2017) xxx–xxx
Fig. 3. Fluorescence microscopic images (scale bar = 5
2.4 M, D).
lm) of A2780 cells after 24 h untreated (A) and treated with abenquines analogues 3g (2.6 lM, B) and 2i (1.2 lM, C and
l
In conclusion: we have evaluated the cytotoxic activity of four
References
natural abenquines (2a–d) and fourteen synthetic analogues (2e–
j; 3a–h) against six human tumor cell lines and non-malignant
mouse fibroblasts (NIH 3T3) using SBR assay. From the data, the
natural abenquines (2a–d) and their analogues having benzoyl
group with an amino acid group (3a–d) did not show cytotoxic
activity of interest. However, the analogues carrying an
ethylpyrimidinyl or an ethylpyrrolidinyl group, with either the
acetyl group (2h–j) or benzoyl group (3f–g) proved to be the most
active compounds, showing EC50 values ranging between 0.6 and
1. Godfray HCJ, Beddington JR, Crute IR, et al. Science (80). 2010;327:812–818.
2. Newsome R. Food Technol. 2010;64:48–56.
3. Varmus H, Trimble EL. Science (80). 2011;3:1–4.
4. World Health Organization. 2014.
5. Varmus H, Kumar HS. Science (80). 2013;5:1–6.
6. Harvey AL, Edrada-Ebel R, Quinn RJ. Nat Rev Drug Discov. 2015;14:111–129.
7. Newman DJ, Cragg GM. J Nat Prod. 2016;79:629–661.
8. El-Najjar N, Gali-Muhtasib H, Ketola R, Vuorela P, Urtti A, Vuorela H. Phytochem
Rev. 2011;10:353–370.
9. Abraham I, Joshi R, Pardasani P, Pardasani RT.
2011;22:385–421.
J Braz Chem Soc.
3.4 lM. Moreover, as shown by an additional AO/PI staining exper-
10. Donohoe TJ, Jones CR, Barbosa LCA. J Am Chem Soc. 2011;133:16418–16421.
11. Donohoe TJ, Jones CR, Kornahrens AF, et al. J Org Chem. 2013;78:12338–12350.
12. Zhang M, Gaisser S, Nur-E-Alam M, et al. J Med Chem. 2008;51:5494–5497.
13. Miyata Y. Curr Pharm Des. 2005;11:1131–1138.
14. Tomasz M. Chem Biol. 1995;2:575–579.
15. Danshiitsoodol N, de Pinho CA, Matoba Y, Kumagai T, Sugiyama M. J Mol Biol.
2006;360:398–408.
16. Phillips RM, Hendriks HR, Peters GJ. Br J Pharmacol. 2013;168:11–18.
17. Schulz D, Beese P, Ohlendorf B, et al. Antibiot (Tokyo). 2011;64:763–768.
18. Nain-Perez A, Barbosa LCA, Maltha CRA, Forlani G. Tetrahedron Lett.
2016;57:1811–1814.
iment, it was found that compounds 2i and 3g, were unable to kill
cells by a ‘‘classical” apoptosis pathway, other mechanisms may be
involved in cytotoxicity of these compounds. Further intensive
modifications in the core of benzoquinone group of the abenquine
and studies concerning the mode of action of these compounds are
currently performed in our laboratory, and the results will be
reported in due course.
19. Pereira UA, Barbosa LCA, Maltha CRA, Demuner AJ, Masood MA, Pimenta AL.
Bioorg Med Chem Lett. 2014;24:1052–1056.
Acknowledgments
20. Wiemann J, Heller L, Csuk R. Bioorg Med Chem Lett. 2016;26:907–909.
21. Rodríguez-Hernández D, Demuner AJ, Barbosa LCA, Heller L, Csuk R. Eur J Med
Chem. 2016;115:257–267.
22. Rodríguez-Hernández D, Demuner AJ, Barbosa LCA, Csuk R, Heller L. Eur J Med
Chem. 2015;105:57–62.
23. Pereira U, Altoe T, Barbosa LCA, et al. Chem Commun. 2016;7:345–352.
24. Nithianandam VS, Erhan S. Polymer. 1998;39:4095–4098.
25. Barbosa LCA, Pereira UA, Maltha CRA, et al. Molecules. 2010;15:5629–5643.
We are grateful to the following Brazilian agencies: Conselho
Nacional de Desenvolvimento Científico e Tecnológico (CNPq, grant
400746-2014) for research fellowships (LCAB). Fundação de
Amparo e Pesquisa de Minas Gerais (FAPEMIG, grant APQ1557-
15) for research fellowship (ANP) and CAPES-PNPD for research fel-
lowship (DRH). The cell lines were kindly provided by Dr. Thomas
Müller (Dept. of Haematology/Oncology, Martin-Luther University
Halle-Wittenberg). Support by the ‘‘Science Campus- Plant based
bioeconomy” is gratefully.
26. Skehan P, Storeng R, Scudiero D, et al.
1990;82:1107–1112.
J
National Cancer Inst.
27. Daletos G, de Voogd NJ, Müller WEG, et al. J Nat Prod. 2014;77:218–226.
28. Delgado V, Ibacache A, Theoduloz C, Valderrama JA. Molecules.
2012;17:7042–7056.
29. Hawas UW, Shaaban M, Shaaban Ka, et al. J Nat Prod. 2009;72:2120–2124.
30. Siewert B, Pianowski E, Csuk R. Eur J Med Chem. 2013;70:1–14.
31. Weber GF. Molecular Therapies of Cancer. Springer; 2015.
A. Supplementary material
Supplementary data associated with this article can be found, in
the online version, at http://dx.doi.org/10.1016/j.bmcl.2017.01.079.
Please cite this article in press as: Nain-Perez A., et al. Bioorg. Med. Chem. Lett. (2017), http://dx.doi.org/10.1016/j.bmcl.2017.01.079