Target synthesis of bioactive thioglycolurils, based on QSAR predictions

Article abstract of DOI:10.1016/j.mencom.2013.07.007

Target synthesis of substituted thioglycolurils possessing sedative activity predicted by preliminary 3D-QSAR simulation of bioactive structures has been carried out.

Full text of DOI:10.1016/j.mencom.2013.07.007

Mendeleev
Communications
Mendeleev Commun., 2013, 23, 202–203
Target synthesis of bioactive thioglycolurils, based on QSAR predictions
Galina A. Gazieva,*a Yurii B. Vikharev,b Lada V. Anikina,b Tatyana B. Karpova,a
Angelina N. Kravchenko,a Evgenii A. Permyakova and Igor V. Svitankoa,c,d
a N. D. Zelinsky Institute of Organic Chemistry, Russian Academy of Sciences, 119991 Moscow,
Russian Federation. Fax: +7 499 135 5328; e-mail: gaz@ioc.ac.ru
b Institute of Technical Chemistry, Ural Branch of the Russian Academy of Sciences, 614013 Perm,
Russian Federation
c Department of Chemistry, M. V. Lomonosov Moscow State University, 119991 Moscow, Russian Federation
d Higher Chemical College, Russian Academy of Sciences, 125047 Moscow, Russian Federation.
E-mail: svitanko@mail.ru
DOI: 10.1016/j.mencom.2013.07.007
Target synthesis of substituted thioglycolurils possessing sedative activity predicted by preliminary 3D-QSAR simulation of bioactive
structures has been carried out.
Target synthesis of compounds with predesigned properties is
among the most important tasks of current medicinal chemistry.
A rather popular approach to perform this task involves applying
computational methods for simulating structures with predesigned
bioactivity, followed by synthesis of promising compounds and
experimental confirmation of the pre-computed activity.
For these compounds, geometrical structure computation with
PES optimization was performed, and the points corresponding
to local ‘vertices’ and ‘cavities’ were identified. Subsequently,
these points were divided into classes based on the electrostatic
potential value and on whether each point belonged to a PES
vertex or cavity. All possible pairs of key points were determined
for each structure and then grouped into sets comprising similar
classes of points and similar distances between them. The number
of groups in the classification was based on the distribution of
values to determine the boundaries of dense value groups (the
boundaries were drawn over the places where the density of values
was smaller than in other places). In this way, a set of pairs was
obtained for each structure.
Building the model involved a search for such pairs of points
that were repeated in the molecules of the training set that showed
sedative activity but not in the molecules of inactive compounds.
In such a way, ten pairs of points were selected which most
accurately described the sedative activity of the training set,
i.e., they were descriptors of the structures with respect to the
given property.
Previously,1–4 we have shown that preliminary computation
of the structures of potential bioactive compounds can be enabled
by optimization of the free energy of their bonding with biological
targets. This study deals with glycolurils, i.e., bicyclic bisureas
of 2,4,6,8-tetraazabicyclo[3.3.0]octane series, which currently lack
reported therapeutic targets for biological activity assessment.
The classical 3D-QSAR is the only method for bioactivity com-
putation if a reliable target is unavailable.
The aim of this work was discovery of novel sedative agents
of the thioglycolurils series. We used a 3D-QSAR version involvin
g
quantum-chemical computation of the potential energy surface
(PES) of molecules belonging to the training set, determination
of the extrema of these surfaces, followed by comparison of the
extremum points (‘key points’).†,5,6 A model consisting of common
points of the PES for structures of the compounds belonging
to the training set allows one to estimate the prospects of certain
structures for target synthesis.7,8
In order to check the model efficiency, we constructed the
structures of 12 thioglycolurils with general formula V unreported
in literature and checked how they matched the model.
The starting training set comprised 33 compounds, viz., mono-
and bicyclic ureas, either definitely active or inactive, belonging
to four types of structures, I–IV.
Ar
R
N
N
N
R5
R1
R1
R1
O
S
R3
R3
R = Me, Et
R3
R4
N
N
H
Ar = Ph, 4-BrC6H4, 3-MeOC6H4,
4-MeOC6H4, 4-O2NC6H4, 2-HOC6H4,
4-HO-3,5-(MeO)2C6H2
N
N
N
N
N
OH
OH
R
O
X
X
O
V
N
N
N
O
R4
R4
R2
R6
R2
R2
As a result, three structures were selected to be synthesized:
1 – the structure best matching the model; 2 – a structure with a
smaller degree of matching; reference structure 3 that gave an
ambiguous agreement with the model.
Compounds 1–3 were prepared using a reaction of 3-thioxo-
perhydroimidazo[4,5-e]-1,2,4-triazin-6-ones 4, 5 with aromatic
aldehydes 6–8 in methanol under HCl catalysis (Scheme 1),
similarly to the technique that we developed previously.9
The structures of the compounds synthesized were confirmed
by elemental analyses, 1H NMR spectroscopy and mass spectro-
metry.‡
I
II
III
R6
R7
N
R1
R1, R2, R5, R6 = H, Me, Et, Pr, Bu,
Ph, CH2OH, Ac
R3, R4 = H, Ph
R7 = Me, CH2CH2OH
X = O, S
HN
N
O
O
O
N
HN
R7
R6
R2
N
IV
Calculations were performed using previously elaborated software.5,6
†
© 2013 Mendeleev Communications. All rights reserved.
– 202 –

Mendeleev Commun., 2013, 23, 202–203
R
male white mice with mass of 22–24 g. All data were processed
H
N
statistically by the Student criterion and compared with data for
the control group. The difference in estimates was considered
reliable at p < 0.05.13
N
NH
O
+
ArCHO
N
N
H
S
Our results demonstrate that the predicted and measured
sedative activity of the selected structures do truly match, and
indicate that compound 1 manifests a pronounced sedative effect
by decreasing the motion and exploratory activity of mice, both
in the ‘open field’ test (a twofold decrease in the number of
crossed squares and a threefold decrease in the number of explored
holes in comparison with control mice) and in the ‘elevated plus
maze’ test (a twofold decrease in the total number of entries in
the maze arms and a threefold decrease in the number of rearings
and head dippingings from open maze arms). Compound 2 affects
somewhat the central nervous system by decreasing the motion
activity and increasing the exploratory activity of mice but only in
the ‘open field’ test. Compound 3 does not affect the behavioral
characteristics of the animals.
R
6 Ar = 4-MeOC6H4
4 R =Me
5 R = Et
7 Ar = 4-HO-3,5-(MeO)2C6H2
8 Ar = 3-MeOC6H4
Ar
N
R
N
N
S
i
O
N
N
H
R
1 R = Et, Ar = 4-MeOC6H4
2 R = Me, Ar = 4-HO-3,5-(MeO)2C6H2
3 R = Et, Ar = 3-MeOC6H4
The reference compound, Mebicar, reduces the number of
crossed squares in open field 1.5-fold and the total number of
entries in maze arms twofold, i.e., it has an inhibitory effect on the
motion behavior of animals. On the other hand, it does not affect
the ‘exploratory’behavior of animals and manifests an anxiolytic
effect by prolonging the time spent by the animals in open maze
arms. In comparison with Mebicar, none of the compounds studied
have an anxiolytic effect in the ‘elevated plus maze’ test.
To conclude, target synthesis of novel sedative thioglycolurils
has been carried out, whose structures were simulated by QSAR
methods.
Scheme 1 Reagents and conditions: i, MeOH, conc. HCl, reflux, 1 h.
Further, we studied the psychopharmacological properties of
compounds 1–3. Mebicar {1,3,4,6-tetramethyltetrahydroimidazo-
[4,5-d]imidazole-2,5(1H,3H)-dione}, a daytime tranquilizer that
manifested not only sedative but also anxiolytic effect,10 was
used as the reference compound.
Me
N
Me
N
O
O
N
N
This work was supported by the Ministry of Education
and Science of the Russian Federation (program ‘Scientific and
Educational Research Personnel of Innovative Russia’, grant
nos. 8431 and 8441) and the Russian Foundation for Basic Research
(grant no. 12-03-01036-a).
Me
Me
Mebicar
The effect of the compounds on the motion and exploratory
activity was estimated in the ‘open field’ test,11 while the anxiolytic
activity was estimated in the ‘elevated plus maze’ test12 in outbred
References
1 T. V. Rakitina, A. A. Zeifman, I. Yu. Titov, I. V. Svitanko, A. V. Lipkin,
V. S. Stroylov, O. V. Stroganov, F. N. Novikov and G. G. Chilov, Mendeleev
Commun., 2012, 22, 287.
All compounds 1–3 gave satisfactory elemental analyses. The 1H NMR
‡
spectra were recorded on a Bruker AM-300 spectrometer (300.13 MHz).
Chemical shifts were measured with reference to the residual protons of
the DMSO-d6 solvent (d 2.50 ppm). High resolution mass spectra (HRMS)
of compound 1 were measured on a Bruker micrOTOF II instrument using
electrospray ionization (ESI). Mass spectra of compounds 2 and 3 were
measured on an MS 30 spectrometer.
(E)-1,3-Diethyl-4-(4-methoxybenzylidene)amino-5-thioxohexahydro-
imidazo[4,5-d]imidazol-2(1H)-one 1: yield 65%, mp 208–210°C (decomp.).
1H NMR (DMSO-d6) d: 1.05 (m, 6H, Me), 3.11–3.40 (m, 4H, NCH2),
3.81 (s, 3H, MeO), 5.48 (d, 1H, CH, J 8.4 Hz), 5.93 (d, 1H, CH, J 8.4
Hz), 7.04 (d, 2H, Ar, J 8.6 Hz), 7.71 (d, 2H, Ar, J 8.6 Hz), 9.15 (s, 1H,
N=CH), 9.86 (s, 1H, NH). HRMS, m/z: 348.1481 [M+H]+ (C16H21N5O2S,
D = 2.3 ppm).
(E)-4-(4-Hydroxy-3,5-dimethoxybenzylidene)amino-1,3-dimethyl-
5-thioxohexahydroimidazo[4,5-d]imidazol-2(1H)-one 2: yield 45%,
mp 228–230°C (decomp.). 1H NMR (DMSO-d6) d: 2.74 (s, 3H, MeN),
2.83 (s, 3H, MeN), 3.80 (s, 6H, MeO), 5.37 (d, 1H, CH, J 8.3 Hz), 5.87
(d, 1H, CH, J 8.3 Hz), 7.06 (s, 2H, Ar), 8.98 (s, 1H, OH), 9.00 (s, 1H,
N=CH), 9.85 (s, 1H, NH). MS, m/z (%): 365 (7) [M+], 306 (36), 241 (15),
179 (100), 164 (39), 127 (47), 125 (43), 112 (24), 101 (26), 98 (29), 77 (31),
68 (40).
2 A. A. Zeifman, I. Yu. Titov, I. V. Svitanko, T. V. Rakitina,, A. V. Lipkin,
V. S. Stroylov, O. V. Stroganov, F. N. Novikov and G. G. Chilov, Mendeleev
Commun., 2012, 22, 73.
3 A.Yu. Sukhorukov, S. O. Andryushkevich, G. G. Chilov, A. A. Zeifman,
I. V. Svitanko and S. L. Ioffe, Mendeleev Commun., 2011, 21, 183.
4 L. V. Romashov, A. A. Zeifman, A. L. Zakharenko, F. N. Novikov, V. S.
Stroilov, O. V. Stroganov, G. G. Chilov, S. N. Khodyreva, O. I. Lavrik,
I. Yu. Titov and I. V. Svitan’ko, Mendeleev Commun., 2012, 22, 15.
5 M. I. Kumskov, I. L. Zyranov and I. V. Svitanko, Pattern Recognit.
Image Anal., 1995, 3, 477.
6 E. I. Prokhorov, L.A. Ponomareva, E.A. Permyakov and M. I. Kumskov,
Pattern Recognit. Image Anal., 2011, 21, 542.
7 I. V. Svitan’ko, I. L. Zyryanov, M. I. Kumskov, L. I. Khmel’nitskii,
L. I. Suvorova, A. N. Kravchenko, T. B. Markova, O. V. Lebedev, G. A.
Orekhova and S. V. Belova, Mendeleev Commun., 1995, 49.
8 I. V. Svitanko, D. A. Devetyarov, D. E. Tcheboukov, M. S. Dolmat,
A. M. Zakharov, S. S. Grigor’eva, V. T. Chichua, L. A. Ponomareva and
M. I. Kumskov, Mendeleev Commun., 2007, 17, 90.
9 G. A. Gazieva, P. A. Poluboyarov, L. D. Popov, N. G. Kolotyrkina,
A. N. Kravchenko and N. N. Makhova, Synthesis, 2012, 44, 3366.
10 M. D. Mashkovskii, Lekarstvennye sredstva (Drugs), Novaya Volna,
Moscow, 2010, vol. 1, p. 86 (in Russian).
(E)-1,3-Diethyl-4-(3-methoxybenzylidene)amino-5-thioxohexahydro-
imidazo[4,5-d]imidazol-2(1H)-one 3: yield 48%, mp 145–147°C (decomp.),
1H NMR (DMSO-d6) d: 1.06 (m, 6H, Me), 3.11–3.40 (m, 4H, NCH2),
3.80 (s, 3H, MeO), 5.50 (d, 1H, CH, J 8.4 Hz), 5.90 (d, 1H, CH, J 8.4 Hz),
7.05 (d, 1H, Ar, J 8.6 Hz), 7.36 (m, 3H, Ar), 9.27 (s, 1H, N=CH), 9.98
(s, 1H, NH). MS, m/z (%): 347 (3) [M+], 288 (14), 214 (26), 181 (18),
155 (32), 154 (14), 141 (19), 140 (100), 127 (10), 119 (23), 112 (63), 104 (17),
83 (16).
11 I. R. Bossier, P. Simon and I. M. Zwolf, Therapie, 1964, 19, 571.
12 I. P. Lapin and S. M. Mirzaev, Byul. Eksperim. Biol. Med., 1996, 3, 355
(in Russian).
13 M. L. Belen’kii, Elementy kolichestvennoi otsenki farmakologicheskogo
effekta (Elements of Quantitative Estimation of Pharmacological Effect),
2nd edn., Meditsinskaya Literatura, Leningrad, 1963 (in Russian).
Received: 27th February 2013; Com. 13/4077
– 203 –