Efficient synthesis and cytotoxic activity of some symmetrical disulfides derived from the quinolin-4(1H)-one skeleton ww

Article abstract of DOI:10.1002/ejoc.200900428

The preparation of novel organic disulfides containing the 2(substituted phenyl)quinolin-4(1H)-one ring is described. The synthesis starts from thioanthranilic acid esterified with various bromoacetophenones. Cyclization of the resulting phenacyl thioanth

Full text of DOI:10.1002/ejoc.200900428

SHORT COMMUNICATION
DOI: 10.1002/ejoc.200900428
Efficient Synthesis and Cytotoxic Activity of Some Symmetrical Disulfides
Derived from the Quinolin-4(1H)-one Skeleton
[a]
[a]
[a]
[b]
ˇ
Miroslav Soural,* Jan Hlavác, Pavel Hradil, and Marián Hajdúch
Keywords: Heterocycles / Quinolinone / Biological activity / Cytotoxicity / Rearrangement
The preparation of novel organic disulfides containing the 2-
(substituted phenyl)quinolin-4(1H)-one ring is described.
The synthesis starts from thioanthranilic acid esterified with
various bromoacetophenones. Cyclization of the resulting
phenacyl thioanthranilates in trifluoroacetic acid afforded a
mixture of 2-(substituted phenyl)-3-sulfanylquinolin-4(1H)-
ones and 3,3Ј-disulfanediylbis[2-(substituted phenyl)quin-
disulfanediylbis[2-(substituted phenyl)quinolin-4(1H)-ones]
of high purity. The disulfides exhibited a significant in vitro
cytotoxicity against various cancer cell lines including poly-
resistant subclones. The data obtained are reported and dis-
cussed.
(© Wiley-VCH Verlag GmbH & Co. KGaA, 69451 Weinheim,
olin-4(1H)-ones]. Heating of the mixture in o-xylene gave 3,3Ј- Germany, 2009)
The compounds prepared according to this method were
Introduction
screened for their cytotoxicity by using the in vitro MTT
test, and a significant activity of the hydroxyquinol-
inones[6–8] was detected, whereas 2-substituted 3-amino-
quinolin-4(1H)-ones were inactive.
Flavones and flavonols represent a well-known group of
natural compounds that exhibit interesting biological activi-
ties of various kinds.[1] In contrast, their aza analogues, i.e
derivatives containing the 3-hydroxyquinolin-4(1H)-one
skeleton (termed “hydroxyquinolinones” hereafter), have
not been studied so extensively due to their relative syn-
thetic unavailability. Our department developed a new syn-
thetic approach that allows the versatile and simple prepa-
ration of hydroxyquinolinones[2–4] as well as their 3-amino
analogues[5] by taking advantage of the rearrangement of
anthranilic acid derivatives (Scheme 1).
In the present study, our main goal was to extend our
knowledge of the described rearrangement applicability.
Considering the course of the cyclization of the anthranilic
acid esters/amides we decided to test whether the same kind
of rearrangement would take place for thioanthranilic acid
esters. Theoretically, the products of such cyclizations
should be derivatives of 2-phenyl-3-sulfanylquinolin-4(1H)-
one (termed “sulfanylquinolinones” hereafter) that would
give, after comparison with analogous hydroxyquinolinones
and aminoquinolinones, useful information about struc-
ture/activity relationship in terms of position 3 of the
4(1H)-quinolinone moiety.
Results and Discussion
Synthesis
Scheme 1. General preparation of hydroxyquinolinones and amino-
quinolinones.
Thioanthranilic acid (1) was synthesized from isatoic an-
hydride and hydrogen disulfide according to a published
protocol.[9] We observed that it was not necessary to run
this reaction for 24 h as described by those authors; 2 h
were sufficient for the conversion of isatoic anhydride to
thioanthranilic acid, and the continuous addition of hydro-
gen disulfide could be stopped. The isolated thioanthranilic
acid was esterified with a set of bromoacetophenones con-
taining different substituents in various positions. The cor-
responding phenacyl thioanthranilates 2 were obtained.
When refluxed in trifluoroacetic acid, they cyclized after a
[a] Department of Organic Chemistry, Faculty of Science, Palacký
University,
Trˇ. Svobody 8, 77146 Olomouc, Czech Republic
Fax: +42-0585634405
E-mail: orgchem@aix.upol.cz
[b] Laboratory of Experimental Medicine, Departments of Pediat-
rics and Oncology, Faculty of Medicine, Palacký University and
Faculty Hospital in Olomouc,
Pusˇkinova 6, 77520 Olomouc, Czech Republic
Supporting information for this article is available on the
WWW under http://dx.doi.org/10.1002/ejoc.200900428.
Eur. J. Org. Chem. 2009, 3867–3870
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
3867

M. Soural, J. Hlavácˇ, P. Hradil, M. Hajdúch
SHORT COMMUNICATION
short time to the expected sulfanylquinolinones 3. However, needed for the rearrangement. When disulfides 4b, 4e, 4h,
the products were always accompanied by 40–50% of the 4k, 4o were analyzed by 1H NMR spectroscopy, we ob-
corresponding 3,3Ј-disulfanediylbis[2-(substituted phenyl)- served an interesting anisochronity of the hydrogen atoms
quinolin-4(1H)-ones] (4) (termed “disulfides” hereafter). of the individual quinolinone skeletons. Since the effect was
We tried to suppress this unwanted oxidation reaction by observed only for the derivatives with ortho substitution of
using different cyclizing agents (such as formic acid and the 2-phenyl moiety, we suppose this fact to be caused by
polyphosphoric acid) and by carrying out the reaction in the hindered S–S bond rotation resulting in atropisomer-
degassed TFA under argon, but formation of disulfides per- ism.
sisted in each case. The only exception was the unsubsti-
As the cyclization was not generally usable for the prepa-
tuted phenacyl thioanthranilate (2a) which, after cyclization ration of sulfanylquinolinones 3 but for effective synthesis
under the inert gas, afforded the pure sulfanylquinolinone of disulfides 4, we turned our attention to a possible re-
3a. Interestingly, the resulting ratio of disulfides in the mix- duction of the disulfidic bond, which would afford the re-
tures obtained was not significantly increased when the cy- quired sulfanyl derivatives 3. We chose derivative 4a as a
clization in TFA was carried out for a longer period. The starting material and the only available pure sulfanylquinol-
quantitative oxidation to disulfides 4 was then performed inone 3a as a reference sample for the reduction evaluation.
by briefly refluxing the mixtures in o-xylene after previous After an unsuccessful attempt with sodium borohydride in
evaporation of trifluoroacetic acid. After cooling to room dimethylformamide, the reduction with zinc in acetic acid
temperature, the disulfides 4 precipitated from o-xylene in was tested. In this case the reduction took place, but we
excellent purity without the need of further purification. observed the absence of the NMR signal for the sulfanyl
1
Starting from isatoic anhydride, the reaction sequence de- group in the H NMR spectrum of sulfanylquinolinone 3a
scribed represents a quite simple 4-step route to disulfides (δ = 5.52 ppm) indicating probable formation of a stable
4, which can be accomplished within 1 d. We also found Zn2+ complex. Unfortunately, the attempt to liberate the
that it was possible to shorten the sequence by cyclizing sulfanyl group by acidic hydrolysis with degassed TFA
phenacyl anthranilates 2 to disulfides 4 directly by heating caused partial back-oxidation to disulfide 4a. As another
in o-xylene, but using this procedure the yields were sub- possible synthetic use of the disulfides the oxidative destruc-
stantially lower than precyclization in trifluoroacetic acid. tion of the disulfidic bond was tested. Using unsubstituted
Unfortunately, the latter alternative also could not be ap- disulfide 4a and oxidation with hydrogen peroxide in formic
plied for the preparation of sulfanylquinolinones 3, because acid we smoothly prepared 4-oxo-2-phenyl-1,4-dihydroqu-
using degassed o-xylene and the inert gas led only to the inoline-3-sulfonic acid (5), which was subsequently trans-
corresponding disulfides 4 again (Scheme 2). The above de- formed via its chloride into the model sulfonamide deriva-
scribed results indicate that the formation of disulfides was tive 6. This example represents a possible extension of the
not influenced only by the presence of oxygen in the reac- disulfide chemistry to another group of novel 4-quinolinone
tion mixtures but might be caused by harsh conditions derivatives with potential biological activity (Scheme 3).
Scheme 2. Preparation and cyclization reactions of phenacyl thioanthranilates.
3868
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 3867–3870

Synthesis and Cytotoxic Activity of Some Symmetrical Disulfides
exhibited the same cytotoxicity against CEM. Derivatives
2n and 4n were evaluated as the most active ones from each
group. This confirms our previous promising experience
with such substituted hydroxyquinolinones.[6] Derivative 4n
exhibited even submicromolar activity except against the
chemotherapy-resistant subclones CEM-DNRB and K562-
Tax.[11]
The cytotoxicity of the sulfonamide 6 and of the single
successfully prepared sulfanylquinoline 3a has also been
demonstrated. Comparison of 3a with the corresponding
disulfide 4a shows a 2.5–15 times lower cytotoxicity against
all tested cell lines (Table 2).
Table 2. Cytotoxicity results for compounds 3a and 6 (IC50 [µ]).[a]
CEM
CEM-DNRb
K562
K562-Tax
A549
3a
6
11.0
201.5
9.7
115.2
8.2
154.5
13.2
134.0
3.8
190.0
[a] Average values of IC50 from 3 to 4 independent experiments
with the standard deviation ranging from 10 to 25% of the average
values.
Scheme 3. Chemical transformations of disulfide 4a.
Biological Activity
All synthesized phenacyl thioanthranilates 2 and disul-
fides 4 were tested for their in vitro cytotoxic activity
against various human malignant cell lines by using the
MTT test.[10] Phenacyl thioanthranilates 2 did not exhibit
any significant activity (IC50 Ͼ 36 µ) except derivative 2n
with medium cytotoxicity against CEM (IC50 = 11.3 µ).
The general cytotoxicity of disulfides 4 was substantially
higher, in the range IC50 = 0.73–33.1 µ (Table 1). In most
cases, except for the diametrically different cytotoxicity of
derivatives 4k and 4n, the tested substituents as well as their
position on the 2-phenyl moiety did not dramatically influ-
ence the activity/selectivity. Although comparison of the
cytotoxicity results for derivatives 2 and 4 generally shows
the manifest necessity of the presence of the quinolinone
moiety, one exception was detected: derivatives 2k and 4k
Conclusions
Our general synthesis of sulfanylquinolinones using the
rearrangement of phenacyl thioanthranilates was unsuc-
cessful. However, the disulfides obtained as unwanted reac-
tion products were identified as an interesting group of de-
rivatives with significant activity against human malignant
cell lines including chemotherapy-resistant subclones. One
example allows the prediction that the disulfides may exert
an even higher cytotoxicity compared to the sulfanylquinol-
inones that were the original target of the research. In ad-
dition, the prepared disulfides can serve as a starting mate-
rial for the synthesis of various novel derivatives of 4(1H)-
quinolinone, as an example a synthetic route leading to sul-
fonamides was demonstrated. The disulfides were active
against childhood acute T-lymphoblastic leukemia cells and
thus form a new group of lead compounds for further stud-
ies as potential agents in this orphan disease.
Table 1. Cytotoxicity results for compounds 4 (IC50 [µ]).[a]
CEM
CEM-DNRB
K562
K562-Tax
A549
4a
4b
4c
4d
4e
4f
4g
4h
4i
4j
4k
4l
4m
4n
4o
4p
4r
4s
2.6
8.2
2.9
3.8
4.4
2.8
3.5
2.6
2.9
8.2
42.7
9.9
7.8
0.73
2.6
8.8
2.9
4.4
3.4
8.8
6.2
5.3
8.5
11.1
8.4
6.7
7.1
9.9
24.7
7.6
7.4
3.54
9.2
10.2
9.7
5.0
2.3
3.5
2.7
4.5
3.5
2.9
4.1
3.2
3.3
5.3
12.4
7.9
9.1
0.73
3.6
11.2
3.3
3.1
8.2
11.5
9.3
10.9
11.4
18.8
18.8
7.8
11.5
14.3
33.1
12.4
17.2
2.4
14.7
31.7
11.1
9.9
2.7
3.2
5.4
4.7
3.1
5.2
5.2
3.0
3.2
2.7
3.1
4.7
6.1
0.95
3.0
4.0
3.3
4.4
Experimental Section
Chemical Synthesis
General Procedure for the Preparation of Phenacyl Thioanthranil-
ates 2: Thioanthranilic acid (2 g, 13 mmol) was dissolved in DMF
(10 mL), and triethylamine (1.4 mL, 13 mmol) was added. After
stirring at room temperature for 10 min, bromoacetophenone
(13 mmol) was added, and the reaction mixture was stirred for an-
other 60 min. Then ice-cold water was poured into the mixture.
The precipitated material was collected by suction and washed with
water. If the precipitation did not occur, the product was extracted
with diethyl ether, which was subsequently evaporated in vacuo.
Crude products were crystallized from a mixture of ethanol/diethyl
ether to give the corresponding anthranilates as light yellow crys-
tals.
[a] Average values of IC50 from 3 to 4 independent experiments
with the standard deviation ranging from 10 to 25% of the average
values.
General Procedure for the Preparation of Disulfides 4. Method A
(Used for the Synthesis): Phenacyl thioanthranilate 2 (100 mg) was
Eur. J. Org. Chem. 2009, 3867–3870
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
www.eurjoc.org
3869

M. Soural, J. Hlavácˇ, P. Hradil, M. Hajdúch
SHORT COMMUNICATION
refluxed in trifluoroacetic acid (1 mL) for 60 min. Then the tri-
fluoroacetic acid was evaporated in vacuo, and o-xylene (1 mL) was
added. The reaction mixture was refluxed for 30 min and sub-
sequently cooled to room temperature. The precipitated material
was collected by suction, washed with o-xylene, ethanol and dried.
Method B: Phenacyl thioanthranilate 2 (100 mg) was refluxed in o-
xylene (2 mL) for 120 min. After cooling to room temperature, the
precipitated material was collected by suction, washed with o-
xylene, ethanol and dried.
dilutions, in 20-µL aliquots, of the intended test concentration were
added to the microtiter plate wells at time zero. All test compound
concentrations were examined in duplicate. Incubation of the cells
with the test compounds lasted for 72 h at 37 °C, in 5 % CO2 at
100% humidity. At the end of the incubation period, the cells were
assayed by using MTT. Aliquots (10 µL) of the MTT stock solution
were pipetted into each well and incubated for a further 1–4 h.
After this incubation period, the formazan produced was dissolved
by the addition of 100 µL/well of 10% aq. SDS (pH = 5.5), fol-
lowed by a further incubation at 37 °C overnight. The optical den-
sity (OD) was measured at 540 nm with a Labsystem iEMS Reader
MF. Tumour cell survival (TCS) was calculated by using the follow-
ing equation: TCS = (ODdrug–exposed well/mean ODcontrol wells)ϫ
100%. The TCS50 value, the drug concentration lethal to 50% of
the tumour cells, was calculated from appropriate dose-response
curves.
Preparation of 2-Phenyl-3-sulfanylquinolin-4(1H)-one (3a): Phenacyl
thioanthranilate 2a (100 mg) was refluxed in degassed trifluoro-
acetic acid (1 mL) under argon for 120 min. Then the reaction mix-
ture was poured into ice-cold water, and the precipitated material
was collected by suction, washed with water and dried. An analyti-
cal sample was crystallized from ethanol.
Preparation of 4-Oxo-2-phenyl-1,4-dihydroquinoline-3-sulfonic Acid
(5): Disulfide 4a (0.66 g, 1.3 mmol) was dissolved in formic acid
(5 mL), and 25% hydrogen peroxide (1.8 mL, 13 mmol) was added.
After 30 min of stirring at room temperature, the precipitated white
solid was collected by suction and washed with water.
Supporting Information (see footnote on the first page of this arti-
cle): Analytical data of the compounds prepared.
Acknowledgments
Preparation of 2-Phenyl-3-(piperidin-1-ylsulfonyl)quinolin-4(1H)-one
(6): Sulfonic acid 5 (200 mg. 0.6 mmol) was dissolved in pyridine
(0.3 mL), and thionyl chloride (0.6 mL, 8 mmol) was added drop-
wise. The reaction mixture was stirred at 50 °C for 2 h and then
concentrated to dryness. The resulting oily residue was suspended
in pyridine (1 mL), and piperidin (0.75 mmol) was added. The reac-
tion mixture was stirred at 50 °C for 2 h, and the solvents were
evaporated to dryness. The resulting residue was suspended in cold
water (10 mL) and collected by suction. The crude product was
crystallized from ethanol.
This project was supported in part by the Ministry of Education,
Youth and Sport of the Czech Republic (grants MSM6198959216,
LC 07017) and FM EEA/Norska (A/CZ0046/1/0022 and CZ0099).
[1] G. Brahmachari, D. Gorai, Curr. Org. Chem. 2006, 10, 873–
898.
[2] P. Hradil, J. Jirman, Collect. Czech. Chem. Commun. 1995, 60,
1357–1366.
[3] P. Hradil, J. Hlavácˇ, K. Lemr, J. Heterocycl. Chem. 1999, 36,
141.
[4] P. Hradil, L. Kvapil, J. Hlavácˇ, T. Weidlich, A. Lycˇka, K. Lemr,
J. Heterocycl. Chem. 2000, 37, 831.
[5] P. Hradil, M. Grepl, J. Hlavácˇ, M. Soural, M. Malonˇ, V. Berto-
lasi, J. Org. Chem. 2006, 71, 819.
[6] M. Soural, J. Hlavácˇ, P. Hradil, I. Frysˇová, M. Hajdúch, V.
Bertolasi, Eur. J. Med. Chem. 2006, 41, 467–474.
[7] P. Krejcˇí, P. Hradil, J. Hlavácˇ, M. Hajdúch, Patent
WO2008028427, 2008.
[8] P. Hradil, P. Krejcˇí, J. Hlavácˇ, I. Wiedermannová, A. Lycˇka, V.
Bertolasi, J. Heterocycl. Chem. 2004, 41, 375–379.
[9] A. T. Fanning, G. R. Bickford, T. D. Roberts, J. Am. Chem.
Soc. 1972, 94, 8505–8506.
Biological Activity
Cell Lines: All cells were purchased from the American Tissue Cul-
ture Collection (ATCC), unless otherwise indicated. The daunorub-
icin-resistant subline of CEM cells (CEM-DNR bulk) and paclit-
axel-resistant subline K562-tax were selected in our laboratory by
the cultivation of maternal cell lines in increasing concentrations
of daunorubicine or paclitaxel, respectively.[11] The cells were main-
tained in Nunc/Corning 80 cm2 plastic tissue culture flasks and cul-
tured in cell culture medium (DMEM/RPMI 1640 with 5 g/L glu-
cose, 2 m glutamine, 100 U/mL penicillin, 100 µg/mL streptomy-
cin, 10% fetal calf serum, and NaHCO3).
[10] M. Hajdúch, V. Mihal, J. Minarik, E. Faber, M. Safarova,
Cytotechnology 1996, 19, 243–245.
Cytotoxic MTT Assay:[10,11] The cell suspensions were prepared
and diluted according to the particular cell type and the expected
target cell density (2500–30000 cells/well based on cell growth char-
acteristics). The cells were added by pipette (80 µL) into 96-well
microtiter plates. The inoculates were allowed a pre-incubation
period of 24 h at 37 °C and 5% CO2 for stabilisation. Four-fold
[11] V. Nosková, P. Dzˇubák, G. Kuzmina, A. Ludková, D. Stehlík,
ˇ
R. Trojanec, A. Janosˇtáková, G. Korˇínková, V. Mihál, M.
Hajdúch, Neoplasma 2002, 49, 418–425.
Received: April 17, 2009
Published Online: July 1, 2009
3870
www.eurjoc.org
© 2009 Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim
Eur. J. Org. Chem. 2009, 3867–3870