Synthesis of new 4-phenylpyrimidine-2(1H)-thiones and their potency to inhibit COX-1 and COX-2

Article abstract of DOI:10.1016/j.ejmech.2015.07.003

Several new 4-phenylpyrimidine-2(1H)-thiones have been prepared and investigated for their potencies to inhibit COX-1 and COX-2 enzymes, and COX-2 expression in THP-1 cells. Structure-activity-relationships and physicochemical parameters are discussed. Pharmacophore screening and docking studies were carried out for the most active compound.

Full text of DOI:10.1016/j.ejmech.2015.07.003

European Journal of Medicinal Chemistry 101 (2015) 552e559
Contents lists available at ScienceDirect
European Journal of Medicinal Chemistry
journal homepage: http://www.elsevier.com/locate/ejmech
Research paper
Synthesis of new 4-phenylpyrimidine-2(1H)-thiones and their
potency to inhibit COX-1 and COX-2
a
,
*
a
a
a
b
Werner Seebacher , Johanna Faist , Armin Presser , Robert Weis , Robert Saf ,
c
d
c
e
e
Teresa Kaserer , Veronika Temml , Daniela Schuster , Sabine Ortmann , Nadine Otto ,
e
Rudolf Bauer
a
Institute of Pharmaceutical Sciences, Pharmaceutical Chemistry, University of Graz, Universit a€ tsplatz 1, 8010 Graz, Austria
Institute for Chemistry and Technology of Organic Materials (ICTM), Graz University of Technology, Stremayrgasse 9, 8010 Graz, Austria
Institute of Pharmacy, Pharmaceutical Chemistry, Computer-Aided Molecular Design Group, University of Innsbruck, Innrain 80-82, 6020 Innsbruck,
b
c
Austria
d
Institute of Pharmacy, Pharmacognosy, University of Innsbruck, Innrain 80-82, 6020 Innsbruck, Austria
Institute of Pharmaceutical Sciences, Pharmacognosy, University of Graz, Universit a€ tsplatz 4, 8010 Graz, Austria
e
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 1 December 2014
Received in revised form
Several new 4-phenylpyrimidine-2(1H)-thiones have been prepared and investigated for their potencies
to inhibit COX-1 and COX-2 enzymes, and COX-2 expression in THP-1 cells. Structure-activity-
relationships and physicochemical parameters are discussed. Pharmacophore screening and docking
studies were carried out for the most active compound.
29 June 2015
Accepted 2 July 2015
Available online 10 July 2015
©
2015 Elsevier Masson SAS. All rights reserved.
Keywords:
4-Phenylpyrimidine-2(1H)-thiones
Inhibition of cyclooxygenases
1. Introduction
and determined their ability to inhibit COX-1 and COX-2 enzymes,
and COX-2 expression in THP-1 cells.
A single 4-phenylpyrimidine-2(1H)-thione has already been
investigated for its analgesic, anticonvulsant and anti-
inflammatory activities and was found more potent than prednis-
olone as anti-inflammatory agent [1]. Since the anti-inflammatory
2
. Chemistry
The already
investigated
3,4-dihydro-4-methyl-6-
2
activity was described as an inhibition of plasma-PGE and pro-
phenylpyrimidine-2(1H)-thione (2a) can be prepared by the reac-
tion of 4-phenylbut-3-en-2-one (1a) with thiourea [2]. Alterna-
tively, a reaction of 4-phenylbut-3-en-2-one (1a) with ammonium
thiocyanate leads to compound 2a, too [3]. We applied this method
to prepare a series of new pyrimidinethiones 2b-2j with various
substituents at the aromatic moiety and for the synthesis of a single
non-aromatic derivative 2k to investigate the influence of the
substitution patterns on the biological activity. For the preparation
of compounds 2a-2j, 4-phenylbut-3-en-2-one (1a) or substituted
tection against carrageenan-induced oedema, it seemed to be
interesting to investigate whether this activity is related to COX
inhibition. Therefore, we prepared a series of similar compounds
3
Abbreviations: BCS, biopharmaceutics classification system; CH OH, methanol;
COX, cyclooxy-genase; DMSO, dimethylsulfoxide; EDTA, ethylenediaminetetra-
acetic acid; EIA, enzyme immuno assay; ELISA, enzyme linked immunosorbent
assay; GAPDH, glycerinaldehyde-3-phosphate dehydrogenase; GOLD, genetic opti-
mization for ligand docking; LPS, lipopolysaccharide; MR, molar refractivity; MW,
4-arylbut-3-en-2-ones 1b-1j were used as starting materials. The
molecular weight; PDB, protein data base; PGE
2
, prostaglandin E
2
; PGHS, prosta-
latter were prepared by aldol condensation of substituted benzal-
dehydes with acetone following reported procedures [4,5]. The
non-aromatic pyrimidinethione 2k was prepared from 5-
methylhexen-2-one (1k) (Scheme 1). Evidence of structure was
glandin H synthase; PMA, phorbol-12-myristate-13-acetate; QED, quantitative es-
timate of drug-likeness; RMSD, root-mean-square deviation of atomic position;
SAR, structure activity relationship; TLC, thin layer chromatography; TRIS, tris(hy-
droxymethyl)-aminomethane; tPSA, topological polar surface area.
13
achieved by NMR spectroscopy. The signal of C-2 in C NMR
*
Corresponding author.
E-mail address: we.seebacher@uni-graz.at (W. Seebacher).
spectra appears at 173e175 ppm, a typical value for such thiourea
http://dx.doi.org/10.1016/j.ejmech.2015.07.003
223-5234/© 2015 Elsevier Masson SAS. All rights reserved.
0

W. Seebacher et al. / European Journal of Medicinal Chemistry 101 (2015) 552e559
553
Reverse Transcription Kit (Life Technologies). Relative expression
was quantified via real-time PCR with TaqMan probes for NF- B1
NM_003998) and COX-2 (NM_000963) against endogenous con-
trol GAPDH (predesigned TaqMan assay) via DDCt-method [9].
k
(
4. Molecular modeling
The tested compounds were analyzed in-silico with two target-
based methods: Pharmacophore modeling and docking. Pharma-
cophore models are a three-dimensional array of physicochemical
features that represent the interacting functionalities of a given
molecule with its target [10]. These features represent common
interactions such as hydrogen bonds (H-bonds), ionic interactions,
and hydrophobic interactions [11]. In principle, compounds that
map a pharmacophore model, and therefore could interact in a
similar manner as already known ligands, have an increased
probability to be active against a specific target.
In this study, pharmacophore modeling was performed in
LigandScout (version 3.02). Using this program, either a structure-
or ligand-based modeling approach can be employed for the gen-
eration of pharmacophore models. In the structure-based
approach, the interaction patterns of a ligand-target complex are
used to calculate an initial model. In the ligand-based modeling,
common pharmacophore features are extracted from three-
dimensionally aligned known active molecules. These initial
models can then be refined. In this study, all models were calcu-
lated based on the binding mode of co-crystallized inhibitors,
because multiple ligand-target complexes were available in the
Protein Data Bank [12]. If a part of the ligand fulfills the feature-
specific distances, angles, and electronic requirements in relation
to the protein binding site [11], a feature is placed on it. The
resulting and refined 3D pharmacophore model can then be used
for virtual screening. For this purpose, a 3D multi-conformational
library of molecules is generated with OMEGA [13,14]. During vir-
tual screening, LigandScout calculates a pharmacophore for each
conformer and matches it to the query pharmacophore with a
pattern-matching based alignment algorithm [15]. A Kabsch algo-
rithm is used to minimize the feature distances between the query
model and the conformers. Each match is evaluated with a
geometrical fit value, which is calculated based on the RMSD be-
tween the fitting molecule and the pharmacophore model.
Furthermore, molecular docking was used to fit the compounds
into the empty binding pocket of the target. The quality of the
resulting binding poses is evaluated with a so-called docking score,
which estimates the binding free energy [16]. To investigate the
experimentally observed SAR, all measured compounds were
screened against a previously reported pharmacophore model
collection for COX-1 and COX-2 [17]. This process helped to
compare the binding site interactions of the investigated com-
pounds with the interaction patterns of well-known COX-inhibitors
such as indomethacin that have been co-crystallized with the two
COX isoforms. To further elucidate the binding mechanism of the
novel active compound, all investigated structures were docked
into the binding site of COX. The docking poses confirmed the
crucial interactions from the pharmacophore model and also
illustrated, why structurally closely related molecules did not
display any biological activity.
Scheme 1. Preparation of compounds 2ae2j. Reagents and conditions: (a) ammonium
thiocyanate, benzene, cyclohexanol, reflux 4e6 h.
structures [6]. The NH protons appear at about 8.8 and 9.7 ppm in
1
H NMR spectra and show long range couplings in HMBC spectra to
the carbonyl carbon C-2. Furthermore, we observed w-couplings in
COSY spectra of compounds 2 between both NH protons, between
H-1 and H-5 and between H-5 and H-3 (Fig. 1).
3
. COX-inhibition
The assays were performed in 96 well plates with purified
PGHS-1 (COX-1) from rat seminal vesicles and purified PGHS-2
COX-2) from sheep placental cotyledons (both Cayman Chemical
Company, Ann Arbor, MI, USA) as previously described [7,8]. The
concentration of PGE was determined using a competitive PGE
(
2
2
EIA Kit (Enzo Life Sciences, Ann Arbor, MI, USA). NS-398 (Cayman
Chemical Company) and indomethacin (MP Biochemicals) were
used as positive controls.
For gene expression assays monocytic THP-1 cells (ECCAC, Lot
No. 07І001, Sigma Aldrich) were differentiated to macrophages
with 12 nM PMA for 48 h in 24-well plates, followed by incubation
with test compounds, DMSO as calibrator and quercetin or dexa-
methasone as positive controls for NF-kB1 and COX-2 gene
expression inhibition, respectively. Afterwards cells were stimu-
lated with LPS and incubated for another three hours, except the
well treated with DMSO (all Sigma Aldrich). RNA was extracted
with GenElute Mammalian Total RNA Miniprep Kit (Sigma Aldrich)
and reverse transcription was carried out with High Capacity cDNA
5. Physicochemical properties
Physicochemical parameters are today considered to have a
crucial role in the selection process of drug candidates during the
early stages of drug discovery [18,19]. Most orally active drugs have
a set of properties that fall in a defined physical and chemical ‘drug-
like space’ [20e22]. For this reason an assessment of drug-likeness
Fig. 1. W couplings in COSY spectrum of compound 2f, marked as arrows.

554
W. Seebacher et al. / European Journal of Medicinal Chemistry 101 (2015) 552e559
for all synthesized compounds was made.
Table 2
Key physicochemical parameters of the synthesized compounds.
compd
MWa
tPSAa (Å2)
pH 7.4
MRa
Log Pa
Log Sa
BCSb
QEDc
6
. Results and discussion
All prepared compounds were tested for their activity to inhibit
2
2
2c
2
2
2
2
2h
2i
a
b
204.29
238.73
283.19
222.28
272.29
234.32
249.29
249.29
238.73
273.18
170.27
24.06
24.06
24.06
24.06
24.06
33.29
67.2
63.91
68.71
71.53
64.12
69.88
70.37
70.23
70.23
68.71
73.52
53.04
1.96
2.56
2.73
2.10
2.84
1.80
1.90
1.90
2.56
3.17
1.48
ꢀ3.04
ꢀ3.79
ꢀ4.13
ꢀ3.33
ꢀ4.08
ꢀ3.06
ꢀ3.76
ꢀ3.76
ꢀ3.79
ꢀ4.53
ꢀ2.36
I
I
II
I
II
I
I
I
I
II
I
0.470
0.498
0.517
0.485
0.535
0.682
0.647
0.647
0.498
0.514
0.409
COX-1 and COX-2 enzymes. The results are presented in Table 1.
Interestingly, compound 2a does not show any inhibitory activity.
The readily observed anti-inflammatory activity [1] must be the
result of a different mechanism of action. Most of the compounds
do not inhibit COX-1 and COX-2 with the exception of 2f, 2g, and
d
e
f
g
2
h. The 4-methoxy compound 2f and the 4-nitro substituted de-
67.2
rivative 2g inhibit only COX-1 whereas the 2-nitro derivative 2h
shows a distinct inhibition of both cyclooxygenases. For that
reason, 2h was further investigated for its IC50 against COX-2
24.06
24.06
24.06
2
2
j
k
a
The molecular weight (MW), log P, log S, topological polar surface area (tPSA)
(
Fig. 2) with the result, that 2h shows a IC50 of around 50
The parent pyrimidinethione 2a as well as its derivative 2h with
the highest inhibitory activity were investigated for their ability to
inhibit gene expression of COX-2 and NF- B1 in THP-1 cells. The
mМ.
and molar refractivity (MR) were calculated using the ChemAxon software JChem
for Excel 14.9.1500.912 (2014).
b
The Biopharmaceutics Classification System (BCS) is a classification of drug
k
substances according to their solubility and permeability properties.
c
results are presented in Fig. 3. Obviously, both compounds induce
gene expression of COX-2 maybe as a compensation mechanism.
The expression of NF-kB1 is scarecely influenced by compounds 2a
The quantitative estimate of drug-likeness (QED) was performed using a pub-
lished drug-likeness score on a scale of 0 (not drug-like) to 1 (drug-like) based on
our calculated physicochemical properties.
and 2h.
Furthermore, physicochemical properties of derivatives 2ae2k
were calculated to characterize them and detect possible differ-
ences which may be responsible for their differing activities. Some
key physicochemical properties of compounds 2 are listed in
Table 2.
All compounds comply with the drug-like classifiers defined by
Lipinsky et al. [23,24], Veber et al. [25] and Ghose et al. [26]. Their
molecular weight is relatively low (MW in the region of 170e283),
the polar surface areas and molar refractivities are within the range
where oral bioavailability and bloodebrain barrier penetration is
plausible [27,28].
To estimate the bioavailability of all synthesized compounds a
provisional BCS classification based on in-silico solubility (log S) and
permeability (log P) data was made [29]. Also in case of ionisable
molecules the use of computed log D values, which are often
difficult to determine, did not show any improvement of accuracy,
therefore the general application of calculated log P for the provi-
sional BCS classification was recommended [30].
Fig. 2. COX-2: IC50 determination for 2h: IC50~50 mМ
In BCS (Biopharmaceutics Classification System) drug sub-
stances have been classified into one of four categories defined by
Amidon et al. [31,32]. Most of our tested substances were found to
be in BCS class I (high solubility, high permeability). The BCS is
generally regarded to be one of the most significant prognostic
tools created to facilitate drug discovery in recent years [33]. In
addition, we calculated values that indicate the drug-likeness of the
tested compounds using a published drug-likeness score (quanti-
tative estimate of drug-likeness, QED). The QED score represents
the geometric mean of eight individual functions (molecular
Table 1
a
Activities of 2ae2k, expressed as % inhibition at 50
m
g/ml.
COX-2
Comp.
COX-1
SD
SD
2
2
2
2
2
2
2
2
2
2
2
a
b
c
d
e
f
g
h
i
n.a
e
n.a.
n.a.
n.a.
5.73
n.a.
10.31
6.85
50.99
n.a.
7.33
n.a.
e
e
e
10.26
15.43
n.a
0.92
33.23
29.81
53.00
3.06
n.a.
11.38
21.01
e
7.94
e
15.72
15.50
14.47
16.93
36.46
e
9.48
11.59
11.95
e
j
k
10.72
e
n.a.
e
p.c.-1
p.c.-2
41.47
n.t.
11.05
e
n.t.
45.06
e
15.15
a
Values represent the average of two to three determinations, n.a.: not active,
n.t.: not tested. p.c.-1 positive control for COX-1 (indomethacin), p.c.-2 positive
control for COX-2 (NS-398).
Fig. 3. Effect of 2h and 2a on COX-2 and NF-
ml, n ꢁ 3, mean value ± SD).
kB1 gene expression (concentration 50 mg/

W. Seebacher et al. / European Journal of Medicinal Chemistry 101 (2015) 552e559
555
weight, octanol-water partition coefficient, number of hydrogen
bond donors, number of hydrogen bond acceptors, molecular polar
surface area, number of rotatable bonds, number of aromatic rings
and number of structural alerts) which were selected on the basis of
published precedence for their relevance in determining drug
likeness [34]. QED provides an efficient means to quantify and rank
the “druggability” of substances a range from 0 (not drug-like) to 1
(
drug-like). In a study this common approach outperformed the
Lipinsky Rule of Five, the Ghose rules and the simpler scoring
system by Gleeson [35] and further performed slightly better than
the Veber rule [34]. Amazingly, the computed QED score of all
synthesized compounds correlates very well with the observed
COX-inhibition data, particularly with regard to COX-1 (r ¼ 0.871).
Furthermore, there exists a significant correlation between exper-
imental data and tPSA (COX1: r ¼ 0.792 and COX-2: r ¼ 0.748).
One of the pharmacophore models (based on the PDB entry
4
COX [36]) found only the active constituent 2h (Fig. 4). This model
Fig. 5. Docking poses of compounds 2h (green) and 2g (grey). Only compound 2h was
predicted to form an H-bond with Tyr355. (For interpretation of the references to
colour in this figure legend, the reader is referred to the web version of this article.)
consisted of two H-bond acceptor features to Tyr355 and Arg120,
respectively. In the virtual screening, these features were mapped
by the nitro group in ortho-position, which also corresponds to the
results of the docking simulation (Fig. 4). The aromatic ring and the
methyl group mapped the two hydrophobic features of the model.
A detailed analysis of the interaction pattern of the docking
poses revealed that all compounds formed H-bonds with Tyr385
and Ser503, respectively. Compounds 2d, 2e, 2f and 2g were also
predicted to interact with Arg120. However, only the active com-
pound 2h formed an additional H-bond with Tyr355. All other
compounds failed to do so, because they either had no H-bond
acceptor substitution on the aromatic ring, or it was not accessible.
Compound 2g, which differs only in the substitution site of the
nitro group (2g is para-substituted, while 2h is ortho-substituted)
was also inactive. Intriguingly, also 2g was not placed in the binding
site in a manner that allowed for an interaction with Tyr355 (Fig. 5).
A comparison of the binding poses of 2h with the originally co-
crystallized ligand of 4COX, indomethacin, shows that indometh-
acin forms additional interactions with Arg120 and Tyr355 via an
ionic interaction and forms an additional H-bond with Ser530.
Furthermore, it is larger and fills the hydrophobic parts of the
binding pocket more thoroughly (Fig. 6).
Fig. 6. Comparison between the binding of the co-crystallized inhibitor indomethacin
(cyan) and 2h (grey). Chemical interactions between indomethacin and the COX
7
. Conclusion
binding site are shown. The features are color-coded: H-bond e red arrow, hydro-
phobic contact e yellow sphere, ionic interaction e red star. (For interpretation of the
references to colour in this figure legend, the reader is referred to the web version of
this article.)
As a surprising result, only one of the prepared compounds, 3,4-
dihydro-6-methyl-4-(2-nitrophenyl)pyrimidine-2(1H)-thione,
showed distinct inhibitory activity on cyclooxygenases. In addition,
the evaluation of predictive ADMET parameters revealed the syn-
thesized compounds to be generally drug-like and favourable for
further investigations. On the other hand, it can be stated that the
already observed anti-inflammatory effects of the parent com-
pound
of
this
investigation,
3,4-dihydro-6-methyl-4-
phenylpyrimidine-2(1H)-thion, must be the result of a different
mechanism of action than inhibition of cyclooxygenases.
8. Experimental
8.1. Instrumentation and chemicals
Melting points were obtained on a Tottoli melting point appa-
ratus. IR spectra: infrared spectrometer system 2000 FT (Perkin
Elmer). UV/VIS: Lambda 17 UV/VIS-spectrometer (Perkin Elmer)
Fig. 4. Compound 2h mapped into the pharmacophore model 4COX that found it in a
virtual screening. The nitro group was predicted to form interactions with Arg120 and
Tyr355. Chemical features are color-coded: H-bond e red arrow, hydrophobic contact
e yellow sphere. (For interpretation of the references to colour in this figure legend,
the reader is referred to the web version of this article.)
carried out in CH
3
OH solutions. NMR spectra: Varian UnityInova
4
00 (300 K) 5 mm tubes, spectra were acquired in DMSO. Chemical
1
shifts were recorded in parts per million (ppm), for H spectra the

556
W. Seebacher et al. / European Journal of Medicinal Chemistry 101 (2015) 552e559
solvent peak (2.49) was used as internal standard and for 13C
spectra the central resonance line of the DMSO signal was used as
the internal reference (39.7). Abbreviations: ArH, aromatic H; ArC,
8. 2.1. 3. (4RS)-(± )-4-(4-Bromophenyl)-3, 4-dihydro-6-
methylpyrimidine-2(1H)-thione (2c). From the reaction of
1.58 g (E)-4-(4-bromophenyl)but-3-en-2-one 1c (7.0 mmol) with
0.51 g ammonium thiocyanate (6.7 mmol) in the presence of 0.22 g
q
aromatic C; ArC , quaternary aromatic C, Signal multiplicities are
3
abbreviated as follows: s, singlet; d, doublet; t, triplet; q quartet; m,
cyclohexanol (2.1 mmol) in 25 ccm of benzene, 203 mg (11%) of 2c
1
ꢃ
multiplet. Coupling constants (J) are reported in Hertz (Hz). H- and
were yielded as white crystals. R
f
¼ 0.47, M.p. 210 C, IR (KBr):
1
3
1
1
1
13
C-resonances were assigned using H, H- and H, C-correlation
~v ¼ 3181, 2978, 1706, 1567, 1482, 1430, 1321, 1290, 1194, 825,
1
13
ꢀ1
1
spectra (gCOSY, gHSQC, gHMBC optimized on 8 Hz). H- and C-
resonances are numbered as given in the formulae. HR-MS: GCT-
Premier, Waters (EI, 70eV). Thin-layer chromatography (TLC): TLC
plates (Merck, silica gel 60 F254 0.2 mm, 200 ꢂ 200 mm); the
compounds were detected in UV light at 254 nm. Solvents and
782 cm ; UV (CH
NMR (CDCl
3
OH):
l
(logε) ¼ 205 (4.263), 264 (4.127) nm; H
3
,
d
, 400 MHz): 1.70 (s, 3H, CH
3
), 4.71 (s, 1H, 5-H), 4.91 (s,
1H, 4-H), 7.18 (d, J ¼ 8.1 Hz, 2H, aromatic H), 7.56 (d, J ¼ 8.1 Hz, 2H,
13
aromatic H), 8.85 (s, 1H, 3-H), 9.62 (s, 1H, 1-H) ppm. C NMR
(CDCl , 100 MHz): 17.85 (CH ), 54.34 (C-4), 98.87 (C-5), 120.63
(aromatic C ), 128.69 (aromatic C), 131.26 (C-6), 131.65 (aromatic C),
3
,
d
3
reagents were obtained from commercial sources. R
f
values were
q
determined using benzene: chloroform: ethanol ¼ 4:4:1 as eluent.
144.25 (aromatic C
q
), 174.38 (C-2) ppm. HRMS (EIþ): calcd.
þ
C
11
H
11BrN
2
S [M ]: 281.9826; found: 281.9836.
8
8
.2. Syntheses
8
. 2.1. 4. (4RS)-(± )-4-(4-Fluorophenyl)-3, 4-dihydro-6-
methylpyrimidine-2(1H)-thione (2d). From the reaction of
3.5 g (E)-4-(4-fluorophenyl)but-3-en-2-one 1d (82 mmol) with
.0 g ammonium thiocyanate (79 mmol) in the presence of 2.5 g
.2.1. General procedure for the syntheses of 6-aryl or 6-alkyl-3,4-
1
6
dihydro-4-methylpyrimidine-2(1H)-thiones (2a-2j)
The appropriate ketone 1 was refluxed with ammonium thio-
cyanate and cyclohexanol in benzene on a water separator for
3
cyclohexanol (25 mmol) in 35 ccm of benzene, 2.7 g (15%) of 2d
were yielded as white crystals. R
ꢃ
f
¼ 0.45, M.p. 199 C, IR (KBr):
4
e6 h. After cooling to room temperature, the precipitate was
~
v ¼ 3214, 2980, 1706, 1568, 1509, 1453, 1430, 1225, 1194, 1177, 1157,
filtered with suction, washed with ether and ethanol and dissolved
in hot ethanol or hot isopropanol, treated with charcoal and
filtered. Little amount of solvent was removed in vacuo and then
the solution was allowed to stand overnight to complete crystalli-
zation. The products were filtered with suction washed with pure
solvent and dried in vacuo over phosphorous pentaoxide. Some
products were yielded in low yields, by this method. In such cases,
unreacted ketone can be recovered from the filtrate of the first
filtration by a concentration and crystallization process.
ꢀ1
8
51, 832 cm ; UV (CH
3
OH):
, 400 MHz): 1.71 (s, 3H, CH
s, 1H, 4-H), 7.16e7.28 (m, 4H, aromatic H), 8.85 (s, 1H, 3-H), 9.61 (s,
l
(logε) ¼ 207 (4.148), 266 (4.119) nm;
1
H NMR (CDCl
3
,
d
3
), 4.72 (s,1H, 5-H), 4.93
(
13
1
4
H, 1-H) ppm. C NMR (CDCl
3
,
d
, 100 MHz): 17.86 (CH
3
), 54.26 (C-
2
), 99.21 (C-5), 115.51 (d, J(C,F) ¼ 21.4 Hz, aromatic C), 128.53 (d,
3
4
J(C,F) ¼ 8.4 Hz, aromatic C), 131.14 (C-6), 141.18 (d, J(C,F) ¼ 2.7 Hz,
1
aromatic C
q
), 161.64 (d, J(C,F) ¼ 243.0 Hz, aromatic C
q
), 174.24 (C-2)
S [M ]: 222.0627; found:
þ
ppm. HRMS (EIþ): calcd. C11
H11FN
2
2
22.0634.
8
.2.1.5. (4RS)-(±)-3,4-Dihydro-6-methyl-4-[4-(trifluoromethyl)
8
.2.1.1. (4RS)-(±)-3,4-Dihydro-6-methyl-4-phenylpyrimidine-2(1H)-
thione (2a). From the reaction of 12 g (E)-4-phenylbut-3-en-2-one
a (82 mmol) with 6 g ammonium thiocyanate (79 mmol) in the
phenyl]pyrimidine-2(1H)-thione (2e). From the reaction of
3.1 (E)-4-[4-(trifluoromethyl)phenyl]but-3-en-2-one 1e
1
g
1
3
(61 mmol) with 4.5 g ammonium thiocyanate (59 mmol) in the
presence of 1.9 g cyclohexanol (19 mmol) in 25 ccm of benzene,
4.01 g (25%) of 2e were yielded as white crystals. R
presence of 2.5 g cyclohexanol (25 mmol) in 35 ccm of benzene,
9
melting point (203 C) corresponds well with the reported one
3
g (48%) of 2a were yielded as white crystals as described [3]. The
ꢃ
f
¼ 0.46, M.p.
ꢃ
ꢃ
ꢃ
196 C, IR (KBr): ~v ¼ 3208, 1707, 1567, 1479, 1451, 1429, 1334, 1311,
(
198e199 C) [3]. R
f
¼ 0.50, M.p. 203 C, IR (KBr): ~v ¼ 3190, 2981,
ꢀ1
1295, 1195, 1177, 1163, 1117, 1107, 1070, 1018, 840, 785 cm ; UV
1706, 1589, 1572, 1492, 1451, 1431, 1348, 1319, 1293, 1200, 1177, 755,
1
ꢀ
1
(CH
3
OH):
l
(logε) ¼ 207 (4.163), 264 (4.116) nm; H NMR (CDCl
3
, d,
6
96, 642 cm ; UV (CH
3
OH):
l
(logε) ¼ 206 (4.252), 263 (4.058) nm;
1
3
400 MHz): 1.70 (s, 3H, CH ), 4.76 (s, 1H, 5-H), 5.04 (s, 1H, 4-H), 7.45
H NMR (CDCl
3
,
d, 400 MHz): 1.70 (s, 3H, CH
3
), 4.72 (s, 1H, 5-H), 4.91
(
d, J ¼ 8.1 Hz, 2H, aromatic H), 7.73 (d, J ¼ 8.4 Hz, 2H, aromatic H),
(
s, 1H, 4-H), 7.22e7.37 (m, 5H aromatic H), 8.83 (s, 1H, 3-H), 9.57 (s,
13
H, 1-H) ppm. 13C NMR (CDCl
8.93 (s, 1H, 3-H), 9.68 (s, 1H, 1-H) ppm.
3
C NMR (CDCl , d,
1
4
3
,
d
, 100 MHz): 17.86 (CH
), 99.41 (C-5), 126.44, 127.56, 128.76 (aromatic C), 130.88 (C-6),
), 174.32 (C-2) ppm. HRMS (EIþ): calcd.
S [M ]: 204.0721; found: 204.0727.
3
), 55.02 (C-
1
3
00 MHz): 17.85 (CH ), 54.57 (C-4), 98.71 (C-5), 124.43 (d,
1
3
J(C,F) ¼ 264.2 Hz, CF ), 125.80 (d, J(C,F) ¼ 3.4 Hz, aromatic C),
3
144.95 (aromatic C
q
2
þ
127.17 (aromatic C), 128.17 (d, J(C,F) ¼ 31.7 Hz, aromatic C
q
), 131.49
C
11 12 2
H N
(C-6), 149.36 (aromatic C
q
), 174.68 (C-2) ppm. HRMS (EIþ): calcd.
þ
C
12
H F
11 3
N
2
S [M ]: 272.0595; found: 272.0610.
8
. 2.1. 2. (4RS)-(± )-4-(4-Chlorophenyl)-3, 4-dihydro-6-
methylpyrimidine-2(1H)-thione (2b). From the reaction of
.96 g (E)-4-(4-chlorophenyl)but-3-en-2-one 1b (16.4 mmol) with
8.2.1.6. (4RS)-(± )-3,4-Dihydro-4-(4-methoxyphenyl)-6-
methylpyrimidine-2(1H)-thione (2f). From the reaction of 2.89 g (E)-
4-(4-methoxyphenyl)but-3-en-2-one 1f (16.4 mmol) with 6.0 g
ammonium thiocyanate (15.8 mmol) in the presence of 0.5 g
2
1
.2 g ammonium thiocyanate (15.8 mmol) in the presence of 0.5 g
3
cyclohexanol (5 mmol) in 50 ccm of benzene, 483 mg (13%) of 2b
were yielded as white crystals. R
ꢃ
3
f
¼ 0.56, M.p. 211 C, IR (KBr):
cyclohexanol (5 mmol) in 50 ccm of benzene, 322 mg (9%) of 2f
ꢃ
~
v ¼ 3179, 2972, 1707, 1568, 1485, 1451, 1431, 1322, 1292, 1195, 1175,
were yielded as white crystals. R ¼ 0.62, M.p. 184 C, IR (KBr):
f
ꢀ1
1
087, 829, 783 cm ; UV (CH
3
OH):
, 400 MHz): 1.70 (s, 3H, CH
H, 5-H), 4.93 (s, 1H, 4-H), 7.24 (d, J ¼ 8.4 Hz, 2H, aromatic H), 7.42
d, J ¼ 8.1 Hz, 2H, aromatic H), 8.86 (s, 1H, 3-H), 9.62 (s, 1H, 1-H)
l
(logε) ¼ 204 (4.227), 263
~v ¼ 3204, 2985, 1706, 1568, 1512, 1489, 1321, 1305, 1271, 1248, 1196,
1
ꢀ1
(
4.151) nm; H NMR (CDCl
3
,
d
3
), 4.71 (s,
1174, 1032, 831 cm ; UV (CH
3
OH):
l
(logε) ¼ 205 (4.259), 264
1
1
(
(4.158) nm; H NMR (CDCl
3H, OCH
), 4.68 (s, 1H, 5-H), 4.85 (s, 1H, 4-H), 6.90 (d, J ¼ 8.8 Hz, 2H,
aromatic H), 7.14 (d, J ¼ 8.4 Hz, 2H, aromatic H), 8.76 (s, 1H, 3-H),
3 3
, d, 400 MHz): 1.71 (s, 3H, CH ), 3.72 (s,
3
13
ppm. C NMR (CDCl
3
,
d
, 100 MHz): 17.86 (CH
3
), 54.30 (C-4), 98.95
13
(
C-5), 128.35, 128.75 (aromatic C), 131.27 (C-6), 132.12, 143.84 (ar-
9.52 (s, 1H, 1-H) ppm. C NMR (CDCl
54.45 (C-4), 55.35 (OCH ), 99.54 (C-5), 114.08, 127.84 (aromatic C),
130.79 (C-6), 137.10, 158.84 (aromatic C ), 173.97 (C-2) ppm. HRMS
3 3
, d, 100 MHz): 17.88 (CH ),
þ
omatic C
2
q
), 174.38 (C-2) ppm. HRMS (EIþ): calcd. C11
H
11ClN
2
S [M ]:
3
38.0331; found: 238.0329.
q

W. Seebacher et al. / European Journal of Medicinal Chemistry 101 (2015) 552e559
557
þ
(
EIþ): calcd. C12
H
14
N
2
OS [M ]: 234.0827; found: 234.0832.
8.2.1.11. (4RS)-(±)-3,4-Dihydro-4-isopropyl-6-methylpyrimidin-
(1H)-thione (2k). From the reaction of 9.21 g (E)-5-methylhex-3-
2
8
.2.1.7. (4RS)-(±)-3,4-Dihydro-6-methyl-4-(4-nitrophenyl)pyrimi-
en-2-one 1k (82 mmol) with 6.0 g ammonium thiocyanate
dine-2(1H)-thione (2g). From the reaction of 2.73 g (E)-4-(4-
nitrophenyl)but-3-en-2-one 1g (14.3 mmol) with 1.05 g ammo-
nium thiocyanate (13.7 mmol) in the presence of 0.44 g cyclo-
(79 mmol) in the presence of 2.5 g cyclohexanol (25 mmol) in
50 ccm of benzene, 9.45 g (70%) of 2k were yielded as white
3
ꢃ
crystals. R
f
¼ 0.47, M.p. 191 C, IR (KBr): ~v ¼ 3204, 2990, 2956, 1707,
3
ꢀ1
hexanol (4.4 mmol) in 50 ccm of benzene, 208 mg (6%) of 2g were
1579, 1505, 1455, 1433, 1387, 1338, 1229, 1193, 756 cm ; UV
ꢃ
1
yielded as yellow crystals. R
f
¼ 0.38, M.p. 215 C, IR (KBr): ~v ¼ 3188,
(CH
400 MHz): 0.76, 0.77 (2d, J ¼ 7.5 Hz, 6H, CH(CH
1H, CH(CH ), 1.67 (s, 3H, CH ), 3.69 (s, 1H, 4-H), 4.48 (s, 1H, 5-H),
8.29 (s, 1H, 3-H), 9.27 (s, 1H, 1-H) ppm. C NMR (CDCl
3 2
100 MHz): 16.60, 17.03 (CH(CH ), 17.98 (CH ), 34.74 (CH(CH ) ),
3
OH):
l
(logε) ¼ 260 (4.100), 205 (3.951) nm; H NMR (CDCl
3
, d,
2
1
980, 1712, 1607, 1572, 1522, 1490, 1429, 1351, 1320, 1289, 1200,
3 2
)
), 1.56e1.66 (m,
ꢀ
1
175, 851, 817, 750 cm ; UV (CH
3
OH):
, 400 MHz): 1.70 (s, 3H, CH
H, 5-H), 5.10 (s, 1H, 4-H), 7.49 (d, J ¼ 8.8 Hz, 2H, aromatic H), 8.24
l
(logε) ¼ 270 (4.294), 206
3
)
2
3
1
13
(
4.196) nm; H NMR (CDCl
3
,
d
3
), 4.77 (s,
3
, d,
1
3
)
2
3
(
d, J ¼ 8.4 Hz, 2H, aromatic H), 8.98 (s, 1H, 3-H), 9.73 (s, 1H, 1-H)
56.91 (C-4), 95.72 (C-5), 132.57 (C-6), 175.10 (C-2) ppm. HRMS
13
þ
ppm. C NMR (CDCl
3
,
d
, 100 MHz): 17.86 (CH
3
), 54.41 (C-4), 98.29
(EIþ): calcd. C
H
8 14
N
2
S [M ]: 170.0878; found: 170.0892.
(
C-5), 124.19, 127.53 (aromatic C), 131.80 (C-6), 146.89, 152.01 (ar-
omatic C
q
), 174.78 (C-2) ppm. HRMS (EIþ): calcd. C11
H
11
N
3
O
2
S
8.3. Determination of COX-Inhibition
þ
[M ]: 249.0572; found: 249.0584.
The assays were performed in 96 well plates with purified
PGHS-1 (COX-1) from ram seminal vesicles and purified PGHS-2
(COX-2) from sheep placental cotyledons (both Cayman Chemical
Company, Ann Arbor, MI, USA). The samples were dissolved in
8
.2.1.8. (4RS)-(±)-3,4-Dihydro-6-methyl-4-(2-nitrophenyl)pyrimi-
dine-2(1H)-thione (2h). From the reaction of 2.96 g (E)-4-(2-
nitrophenyl)but-3-en-2-one 1h (16.4 mmol) with 1.2 g ammo-
nium thiocyanate (15.8 mmol) in the presence of 0.5 g cyclohexanol
DMSO and were tested at a concentration of 50
mg/ml. The incu-
3
(
5 mmol) in 50 ccm of benzene, 725 mg (20%) of 2h were yielded
bation solution contained 0.1 M TRIS/HCl buffer, 18 mM epineph-
ꢃ
as yellow crystals. R
f
¼ 0.47, M.p. 207 C, IR (KBr): ~v ¼ 3203, 2986,
rinehydrogentartrate, 5
PGHS-2 enzyme and 50
m
M hematin 0.2 units/well PGHS-1 or
1703, 1584, 1527, 1493, 1435, 1350, 1317, 1284, 1274, 1228, 1194, 857,
m
M Na EDTA for the COX-2 assay. 10 l of
m
2
ꢀ
1
1
7
84, 750, 718, 639 cm ; UV (CH
NMR (CDCl , 400 MHz): 1.69 (s, 3H, CH
H, 4-H), 7.50e7.56 (m, 2H, aromatic H), 7.84 (t, J ¼ 7.3 Hz, 1H, ar-
omatic H), 7.98 (d, J ¼ 8.4 Hz, 1H, aromatic H), 8.68 (s, 1H, 3-H), 9.83
s,1H,1-H) ppm. 13C NMR (CDCl
,100 MHz): 17.88 (CH ), 51.29 (C-
), 97.55 (C-5), 124.66, 128.86, 129.10 (aromatic C), 132.17 (C-6),
34.76 (aromatic C), 139.69, 146.47 (aromatic C ), 175.17 (C-2) ppm.
S [M ]: 249.0572; found: 249.0569.
3
OH):
l
(logε) ¼ 268 (3.735) nm; H
the samples or inhibitor/positive control were added and pre-
incubated for 5 min at room temperature. Then 5 M arachidonic
acid dissolved in EtOH p.a. were added to start the reaction. After an
incubation period of 20 min at 37 C the reaction was stopped by
3
,
d
3
), 4.83 (s, 1H, 5-H), 5.42 (s,
m
1
ꢃ
(
4
1
3
,
d
3
adding formic acid 10%. NS-398 (Cayman Chemical Company) and
Indomethacin (MP Biochemicals) were used as positive controls for
the COX-2 and the COX-1 assay respectively. The concentration of
PGE2 was determined by a competitive PGE2 EIA Kit (Enzo Life
Sciences, Ann Arbor, MI, USA) which was used as described by the
manufacturers. The absorbance was measured with the ELISA
reader “rainbow” (TECAN). The absorbance is reversed proportional
to the concentration of PGE2. Inhibition refers to reduction of PGE2
formation in comparison to a blank and is calculated according to
the following equation:
q
þ
HRMS (EIþ): calcd. C11
11 3 2
H N O
8
. 2.1. 9. (4RS)-(± )-4-(2-Chlorophenyl)-3, 4-dihydro-6-
methylpyrimidin-2(1H)-thione (2i). From the reaction of 14.8 g (E)-
-(2-chlorophenyl)but-3-en-2-one 1i (82 mmol) with 6.0
4
g
ammonium thiocyanate (79 mmol) in the presence of 2.5 g cyclo-
hexanol (25 mmol) in 35 ccm of benzene, 10.5 g (56%) of 2i were
yielded as white crystals. R
2
3
ꢃ
¼ 0.56, M.p. 223 C, IR (KBr): ~v ¼ 3165,
f
987, 1704, 1583, 1570, 1498, 1467, 1435, 1384, 1323, 1286, 1269,
inhibition ½%ꢄ ¼ 100 ꢀ ðconc:PGE2 ðsampleÞ
ꢂ 100Þ÷ðconc:PGE2 ðblankÞÞ
ꢀ
1
1
2
(
202,1036, 752, 736, 641 cm ; UV (CH
3
OH):
, 400 MHz): 1.66 (s, 3H, CH
d, J ¼ 1.1 Hz, 1H, 5-H), 5.24 (s, 1H, 4-H), 7.27e7.31 (m, 2H aromatic
H), 7.39e7.42 (m, 2H, aromatic H), 8.80 (s, 1H, 3-H), 9.72 (s, 1H, 1-H)
ppm. 13C NMR (CDCl
, 100 MHz): 17.83 (CH ), 52.59 (C-4), 97.15
C-5), 128.05, 128.13, 129.13, 129.69 (aromatic C), 129.90 (aromatic
l
(logε) ¼ 207 (4.283),
1
64 (4.181) nm; H NMR (CDCl
3
,
d
3
), 4.80
The samples were tested at least two times in duplicate in in-
dividual experiments. For the IC50 determination six different
3
,
d
3
concentrations (4
mMe100 mM) were tested in duplicate in at least
(
three individual experiments.
C
q
), 131.62 (C-6), 141.83 (aromatic C
q
), 175.35 (C-2) ppm. HRMS
For gene expression assay, THP-1 cells were cultivated in 24-
well plates in media supplemented with PMA at a final concen-
tration of 12 nM in order to differentiate the monocytes to mac-
þ
(
EIþ): calcd. C11
H11ClN
2
S [M ]: 238.0331; found: 238.0346.
8
.2.1.10. (4RS)-(± )-4-(2,4-Dichlorophenyl)-3,4-dihydro-6-
methylpyrimidine-2(1H)-thione (2j). From the reaction of 5.6 g (E)-
-(2,4-dichlorophenyl)but-3-en-2-one 1j (26 mmol) with 1.9 g
rophages. After 48 h cells were treated with compounds (50 mg/ml)
and incubated for one hour. One well was treated with 0.1%
dimethylsulfoxide (DMSO) (Sigma Aldrich®) for calibration. Quer-
4
ammonium thiocyanate (25 mmol) in the presence of 0.8 g cyclo-
hexanol (8 mmol) in 12 ccm of benzene, 600 mg (9%) of 2j were
cetin (25 mM) and/or dexamethasone (2.5 nM) were applied as
3
positive controls. After one hour cells were stimulated with Lipo-
ꢃ
®
yielded as white crystals. R
f
¼ 0.58, M.p. 225 C, IR (KBr): ~v ¼ 3168,
polysaccharide (LPS) (O55:B5, 7.5 ng/ml, Sigma Aldrich ) - except
2
8
986, 1704, 1580, 1560, 1490, 1468, 1434, 1385, 1322, 1202, 1105,
the well treated with DMSO e and incubated for another three
hours. RNA isolation: medium was removed and adherent cells
were washed twice with cold PBS in order to remove remaining
monocytes. Extraction of RNA was carried out with Gen Elute™
ꢀ
1
1
47 cm ; UV (CH
, 400 MHz): 1.67 (s, 3H, CH
H, 4-H), 7.26 (d, J ¼ 8.4 Hz, 1H, aromatic H), 7.52 (d, J ¼ 8.4 Hz, 1H,
aromatic H), 7.58 (s, 1H, aromatic H), 8.82 (s, 1H, 3-H), 9.76 (s, 1H, 1-
H) ppm. 13C NMR (CDCl
, 100 MHz): 17.82 (CH ), 52.29 (C-4),
6.68 (C-5), 128.35, 129.11, 129.48 (aromatic C), 130.89 (aromatic
3
OH):
l
(logε) ¼ 205 (4.511), 265 (4.124) nm; H
NMR (CDCl
3
,
d
3
), 4.77 (s, 1H, 5-H), 5.22 (s,
1
®
Mammalian Total RNA Miniprep Kit (Sigma Aldrich ). Lysis solu-
3
,
d
3
tion/2-mercaptoethanol (mixture prepared according to manufac-
turer's instructions) was applied directly onto adherent cells. Lysate
was mixed by pipetting and transferred into a GenElute™ filtration
9
C
q
), 131.96 (C-6), 132.76, 140.90 (aromatic C
q
), 175.31 (C-2) ppm.
þ
HRMS (EIþ): calcd. C11
H10Cl
2
N
2
S [M ]: 271.9942; found: 271.9954.
column. Wells were rinsed with 100 ml lysis solution and the

558
W. Seebacher et al. / European Journal of Medicinal Chemistry 101 (2015) 552e559
solution was added to the filtration column. Subsequent extraction
steps were performed according to manufacturer's protocol. The
purified total RNA was used instantly for reverse transcription and
was afterwards stored at ꢀ80 C. Reverse transcription: total RNA
4787e4792.
[
2] A.-M.A. Sammour, M.M. El-Deen, A.-el-H.M. Nonr, Condensation of unsatu-
rated ketones with thioureas, U.A.R. J. Chem. 13 (1970) 7e24.
3] G. Zigeuner, H. Brunetti, H. Ziegler, M. Bayer, Dihydro-6-methyl-(or -6-styryl-
)-4-phenyl-2(1H)-pyrimidinones (-thiones); Hexahydro-5-benzoyl- (or -cin-
namoyl-)-4,7-diphenyl-2(1H)-quinazolinones (-thiones), Chem. Mon. 101
[
ꢃ
was reversely transcribed into cDNA by means of High Capacity
cDNA Reverse Transcription Kit™ (Life Technologies, Carlsbad, CA,
USA) on a Mastercycler personal (Eppendorf, Hamburg, Germany).
Real-time PCR: quantification of the synthesized cDNA was per-
formed with reagents from Life Technologies: TaqMan™ Universal
PCR Master Mix (no AmpErase® UNG), pre-developed TaqMan™
assay for human Glyceraldehyde 3-phosphate dehydrogenase
(1970) 1767e1787.
[
4] Y. Tamura, Nonsteroidal antiinflammatory agents. 1. 5-Alkoxy-3-
biphenylylacetic acids and related compounds as new potential antiin-
flammatory agents, J. Med. Chem. 20 (1977) 709e714.
[
5] T. Nishimura, Antituberculous compounds. II. Thiosemicarbazones from Ben-
zalacetones, Bull. Chem. Soc. Jpn. 26 (1953) 253e254.
[6] H.O. Kalinowski, S. Berger, S. Braun, C-NMR-Spektroskopie, Georg Thieme
Verlag, Stuttgart, New York, 1984.
7] B.L. Fiebich, M. Grozdeva, S. Hess, M. Hüll, U. Danesch, A. Bodensieck, R. Bauer,
Petasites hybridus extracts in vitro inhibit COX-2 and PGE2 release by direct
interaction with the enzyme and by preventing microglial cells, Planta Med.
13
[
[
[
(
GAPDH), TaqMan™ probes with FAM™ label as well as sequence
detection primers. Each sample was applied and measured in
duplicate in 96-well PCR plates. PCR and fluorescence detection
were performed with 7300 ABI real-time PCR system (Applied
Biosystems, nowadays incorporated into Life Technologies). Gene
expression was relatively quantified against the endogenous con-
trol gene GAPDH via DDCt-method and analysed with SDS software
version 1.2.1.
71 (2005) 12e19.
8] E.A. Reiniger, R. Bauer, Prostaglandin-H-synthase (PGHS)-1 and -2 microtiter
assays for testing of herbal drugs and in vitro inhibition of PGHS-isoenzyms by
polyunsaturated fatty acids from Platycodi radix, Phytomedicine 13 (2006)
164e169.
9] K.J. Livak, T.D. Schmittgen, Analysis of relative gene expression data using
real-time quantitative PCR and the 2-DDCT Method, Methods 25 (2001)
402e408.
[
[
[
[
10] C.G. Wermuth, C.R. Ganellin, P. Lindberg, L.A. Mitscher, Glossary of terms used
in medicinal chemistry (IUPAC recommendations 1998), Pure Appl. Chem. 70
8
.4. Molecular modeling
(1998) 1129e1143.
11] G. Wolber, T. Langer, LigandScout: 3-D pharmacophores derived from
protein-bound ligands and their use as virtual screening filters, J. Chem. Inf.
Model 45 (2005) 160e169.
12] H. Berman, J. Westbrook, Z. Feng, G. Gilliland, T. Bhat, H. Weissig,
I. Shindyalov, P. Bourne, The protein data bank, Nucleic Acids Res. 28 (2000)
All calculations were performed on a multi-core workstation
with 2.4 þ GHz, 8 GB of RAM, and a high-end NVIDIA graphical
processing unit.
235e242.
13] P.C.D. Hawkins, A.G. Skillman, G.L. Warren, B.A. Ellingson, M.T. Stahl,
Conformer generation with OMEGA: algorithm and validation using high
quality structures from the Protein Databank and Cambridge Structural
Database, J. Chem. Inf. Model 50 (2010) 572e584.
8.4.1. Pharmacophore modeling
All compounds and their tautomeric states were screened in
LigandScout 3.03b [11] against a set of previously developed
pharmacophore models described in Temml et al. [17]. 500 con-
formations were calculated for each tautomer of the structures
with OMEGA version 2.3.3 [13,14], implemented in LigandScout.
[
14] P.C.D. Hawkins, A. Nicholls, Conformer generation with OMEGA: learning
from the data set and the analysis of failures, J. Chem. Inf. Model 52 (2012)
2919e2936.
[15] G. Wolber, A.A. Dornhofer, T. Langer, Efficient overlay of small organic mol-
ecules using 3D pharmacophores, J. Comput. Aided Mol. Des. 20 (2006)
7
73e788.
8
.4.2. Docking studies
The crystal structure of methyl flurbiprofen in complex with
[16] D.B. Kitchen, H. Decornez, J.R. Furr, J. Bajorath, Docking and scoring in virtual
screening for drug discovery: methods and applications, Nat. Rev. Drug Dis-
cov. 3 (2004) 935e949.
COX-1 (PDB entry 1HT5, [37]) was employed for the docking studies
with GOLD version 5.1 [38]. For validation of the docking parame-
ters, the co-crystallized ligand was again docked into the binding
site, and the predicted pose retrieved a RMSD value of 1.03 Å.
Subsequently, the whole database of compounds (including
tautomeric states) was docked. The interaction patterns of the
highest ranked docking pose of every compound were manually
analyzed in LigandScout 3.03b.
[
17] V. Temml, T. Kaserer, Z. Kutil, P. Landa, T. Vanek, D. Schuster, Pharmacophore
Modeling for Cyclooxygenase-1 and 2 inhibitors with LigandScout in com-
parison to Discovery Studio, Fut. Med. Chem. 6 (2014) 1869e1881.
[18] H. van de Waterbeemd, E. Gifford, ADMET in silico modeling: towards pre-
diction paradise? Nat. Rev. Drug Discov. 2 (3) (2003) 192e204.
[19] M.S. Lajiness, M. Vieth, J. Erickson, Molecular properties that influence oral
drug-like behavior, Curr. Opin. Drug Discov. Dev. 7 (4) (2004) 470e477.
[
20] M.C. Wenlock, R.P. Austin, P. Barton, A.M. Davis, P.D. Leeson, A comparison of
Physiochemical Property Profiles of development and Marketed oral drugs,
J. Med. Chem. 46 (7) (2003) 1250e1256.
[
21] J.D. Hughes, J. Blagg, D.A. Price, S. Bailey, G.A. DeCrescenzo, R.V. Devraj,
E. Ellsworth, Y.M. Fobian, M.E. Gibbs, R.W. Gilles, N. Greene, E. Huang,
T. Krieger-Burke, J. Loesel, T. Wager, L. Whiteley, Y. Zhang, Physiochemical
drug properties associated with in vivo toxicological outcomes, Bioorg. Med.
Chem. Lett. 18 (17) (2008) 4872e4875.
Acknowledgements
We thank the Austrian Science Fund (FWF) National Research
Network (NFN) Project “Drugs from nature targeting inflammation”
S10711), the foundation “Verein zur F o€ rderung der wissen-
schaftlichen Ausbildung und T a€ tigkeit von Südtirolern an der
Landesuniversit a€ t Innsbruck”, and the Erika Cremer Habilitation
Program of the University of Innsbruck for financial support. We
also thank OpenEye and Inte:Ligand for providing software free of
charge.
[
22] C.A. Lipinski, A. Hopkins, Navigating chemical space for biology and medicine,
Nat. (London, U. K.) 432 (7019) (2004) 855e861.
23] C.A. Lipinski, F. Lombardo, B.W. Dominy, P.J. Feeney, Experimental and
computational approaches to estimate solubility and permeability in drug
discovery and development settings, Adv. Drug Deliv. Rev. 23 (1e3) (1997)
(
[
3
e25.
[
[
24] C.A. Lipinski, Lead- and drug-like compounds: the rule-of-five revolution,
Drug Discov. Today: Technol. 1 (4) (2004) 337e341.
25] D.F. Veber, S.R. Johnson, H.-Y. Cheng, B.R. Smith, K.W. Ward, K.D. Kopple,
Molecular properties that influence the oral bioavailability of drug candidates,
J. Med. Chem. 45 (12) (2002) 2615e2623.
Appendix A. Supplementary data
[26] A.K. Ghose, V.N. Viswanadhan, J.J. Wendoloski, A knowledge-based approach
in designing combinatorial or medicinal chemistry libraries for drug discov-
ery. 1. A qualitative and quantitative characterization of known drug data-
bases, J. Comb. Chem. 1 (1) (1999) 55e68.
[27] H. Pajouhesh, G.R. Lenz, Medicinal chemical properties of successful central
nervous system drugs, NeuroRx 2 (4) (2005) 541e553.
Supplementary data related to this article can be found at http://
dx.doi.org/10.1016/j.ejmech.2015.07.003.
[
28] H. van de Waterbeemd, Physicochemical approaches to drug absorption, in:
R. Mannhold, H. Kubinyi, G. Folkers, H. van de Waterbeemd, B. Testa (Eds.),
Drug Bioavailability, Wiley-VCH Verlag GmbH & Co. KGaA, Weinheim, Ger-
many, 2008, pp. 69e99.
References
[
1] S.A. Said, A. El-G, E. Amr, N.M. Sabry, M.M. Abdalla, Analgesic, anticonvulsant
and ant-inflammatory activities of some synthesized benzodiazepine, tri-
azolopyrimidine and bis-imide derivatives, Eur. J. Med. Chem. 44 (2009)
[29] K.J. Box, J. Comer, Using measured pKa, LogP and solubility to investigate
supersaturation and predict BCS class, Curr. Drug Metab. (9) (2008)
9

W. Seebacher et al. / European Journal of Medicinal Chemistry 101 (2015) 552e559
the chemical beauty of drugs, Nat. Chem. 4 (2) (2012) 90e98.
559
869e878.
[
30] A. Dahan, O. Wolk, Y.H. Kim, C. Ramachandran, G.M. Crippen, T. Takagi,
M. Bermejo, G.L. Amidon, Purely in silico BCS classification: science based
quality standards for the world's drugs, Mol. Pharm. 10 (11) (2013)
[35] M.P. Gleeson, A. Hersey, D. Montanari, J. Overington, Probing the links be-
tween in vitro potency, ADMET and physicochemical parameters, Nat. Rev.
Drug Discov. 10 (3) (2011) 197e208.
4
378e4390.
[36] R.G. Kurumbail, A.M. Stevens, J.K. Gierse, J.J. McDonald, R.A. Stegeman, J.Y. Pak,
D. Gildehaus, J.M. Miyashiro, T.D. Penning, K. Seibert, Structural basis for se-
lective inhibition of cyclooxygenase-2 by anti-inflammatory agents, Nature
384 (1996), 664e648.
[37] B.S. Selinsky, K. Gupta, C.T. Sharkey, P.J. Loll, Structural analysis of NSAID
binding by protaglandine H2 synthase: time-dependent and time-
independent inhibitors elicit identical enzyme conformations, Biochemistry
40 (2001) 5172e5180.
[
[
[
31] G.L. Amidon, H. Lennernaes, V.P. Shah, J.R. Crison, A theoretical basis for a
biopharmaceutic drug classification: the correlation of in vitro drug product
dissolution and in vivo bioavailability, Pharm. Res. 12 (3) (1995) 413e420.
32] A. Dahan, J.M. Miller, G.L. Amidon, Prediction of solubility and permeability
class membership: provisional BCS classification of the world's top oral drugs,
AAPS J. 11 (4) (2009) 740e746.
33] G.L. Amidon, H. Lennernas, V.P. Shah, J.R. Crison, A theoretical basis for a
biopharmaceutic drug classification: the correlation of in vitro drug product
dissolution and in vivo bioavailability, Pharm Res 12, 413-420, 1995-Back-
story of BCS, AAPS J. 16 (5) (2014) 894e898.
[38] G. Jones, P. Willett, R.C. Glen, A.R. Leach, R. Taylor, Development and valida-
tion of a genetic algorithm for flexible docking, J. Mol. Biol. 267 (1997)
727e748.
[34] G.R. Bickerton, G.V. Paolini, J. Besnard, S. Muresan, A.L. Hopkins, Quantifying