Benzanilides with spasmolytic activity: Chemistry, pharmacology, and SAR

Article abstract of DOI:10.1016/j.bmc.2008.04.057

The following study describes the synthesis of new benzanilide derivatives and their pharmacological investigation on smooth muscle preparations of guinea pigs. All compounds were synthesized in good yields and showed a spasmolytic activity without significant effect on vascular smooth muscles and heart muscle preparations. Moreover, further pharmacological investigations as well as in silico studies were performed to elucidate the mechanism of action. Compound 3 showed the most potent spasmolytic activity with an IC₅₀ of 3.25 μM.

Full text of DOI:10.1016/j.bmc.2008.04.057

Available online at www.sciencedirect.com
Bioorganic & Medicinal Chemistry 16 (2008) 5974–5981
Benzanilides with spasmolytic activity:
Chemistry, pharmacology, and SAR
Gerda Brunhofer,a Norbert Handler,a Klaus Leisser,a
Christian R. Studenikb and Thomas Erkera,*
aDepartment of Pharmaceutical Chemistry, University of Vienna, A-1090 Vienna, Althanstrasse 14, Austria
bDepartment of Pharmacology and Toxicology, University of Vienna, A-1090 Vienna, Althanstrasse 14, Austria
Received 18 January 2008; revised 16 April 2008; accepted 23 April 2008
Available online 26 April 2008
Abstract—The following study describes the synthesis of new benzanilide derivatives and their pharmacological investigation on
smooth muscle preparations of guinea pigs. All compounds were synthesized in good yields and showed a spasmolytic activity with-
out significant effect on vascular smooth muscles and heart muscle preparations. Moreover, further pharmacological investigations
as well as in silico studies were performed to elucidate the mechanism of action. Compound 3 showed the most potent spasmolytic
activity with an IC50 of 3.25 lM.
Ó 2008 Elsevier Ltd. All rights reserved.
1. Introduction
colleagues synthesized benzimidazole derivatives as stil-
bene isosteres and proved their spasmolytic activity by
using the rat ileum test.5 In a previous study we investi-
gated the spasmolytic activity of resveratrol that shows
an IC50 of 46.4 lM.6 Therefore, starting point of this
study was also the modulation of the stilbene scaffold.
The new synthesized substances exhibit an increase of
the spasmolytic activity. We replaced the ethylene linker
by a carboxamide group getting a benzanilide structure
also widely known as a very active principle. We tested
the effect of our novel molecules on the terminal ileum of
the guinea pig. Further pharmacological in vitro exper-
iments were undertaken concerning the underlying
modes of action as well as the effect of the substances
on the vasculature. In the following study we present
new substances with spasmolytic activity, which could,
indeed, be beneficial in the treatment of gastrointestinal
disorders.
Gastrointestinal hypermotility and abdominal pain are
symptoms of the irritable bowel syndrome (IBS) and
other functional bowel disorders.1 Unfortunately, the
basis of these gastrointestinal motility disorders which
induce among others abnormally stimulated motility is
still not understood. It is our attempt to identify a new
class of compounds exhibiting spasmolytic activity and
influencing the gastrointestinal hypermotility and over-
action of the intestinal tract without having strong
side-effects like the standard anti-diarrhoeal agent lop-
eramide, which influences significantly the central ner-
vous system. According to the international literature,
stilbenoids show a highly active principle. Gigantol,
lusianthridin, and batatasin III are stilbenoids which
showed a higher than or comparable to the spasmolytic
activity of papaverine at the rat ileum.2,3 Resveratrol
(3,40,5-trihydroxy-trans-stilbene) is a phytoalexin mainly
found in the skin of grapes and in red wine. This natural
product displays a wide range of pharmacological-effects
like for example cardiovascular protective and cancer
chemopreventive properties.4 Navarette-Vazquez and
2. Chemistry
A series of new benzanilide derivatives were synthesized
by using a standard chemical methodology. The reaction
between the appropriate benzoylchloride and the corre-
sponding substituted aniline yielded the desired benzani-
lide derivative. The reaction was carried out at room
temperature and afforded all compounds in a good yield
(Table 1).
Keywords: Benzanilide; SAR; Spasmolytic activity; Smooth muscle
preparations.
*
Corresponding author. Tel.: +43 1 4277 55003; fax: +43 1 4277
9551; e-mail: thomas.erker@univie.ac.at
0968-0896/$ - see front matter Ó 2008 Elsevier Ltd. All rights reserved.
doi:10.1016/j.bmc.2008.04.057

G. Brunhofer et al. / Bioorg. Med. Chem. 16 (2008) 5974–5981
Table 1. Structures of the synthesized benzanilide derivatives 1–10
R1
5975
R2
R4
H
N
R5
R6
R3
O
R8
R7
Compound
R1
R2
R3
R4
R5
R6
R7
R8
1
2
–H
–F
–F
–H
–H
–F
–F
–H
–H
–H
–H
–F
–F
–H
–H
–F
–H
–F
–H
–F
–H
–H
–H
–H
–H
–F
–H
–H
–F
–OCH3
–H
–H
–OCH3
–SCH3
–N(CH3)2
–H
–OCH3
–H
–H
3
–F
–F
–H
–H
–F
–F
4
5
–H
–F
–OCF3
–H
–H
–H
–F
–OCH3
–H
–H
–OCH3
–H
–H
6
–F
–F
7
–F
–F
–F
–F
8
–H
–F
–H
–F
–F
–H
–H
–F
9
10
–H
–NO2
–H
–NO2
–H
–H
–F
–H
3. Pharmacological results and discussion
0
-25
The inotropic and chronotropic effects of the com-
pounds were studied on isolated heart muscle prepara-
tions of guinea pigs in a concentration range between
0.1 and 100 lM. None of the derivatives showed a sig-
nificant change in inotropy or chronotropy up to a con-
centration of 100 lM (data not shown). The relaxing
effect of the compounds was also studied on aortic rings
and terminal ilea. The preparations were stimulated with
90 mM KCl (aortic rings) and 60 mM KCl (terminal
ilea). There was no significant effect of all compounds
on vascular smooth muscles up to a concentration of
100 lM. In terminal ilea all compounds concentration-
dependently relaxed the KCl-induced contraction force
(fc). At the highest concentration studied (100 lM for
compounds 1, 2, 4, 10, and 30 lM for compound 3) 1 re-
duced fc to 28.35 5.67% (from 11.46 1.33 to
3.76 1.30 mN, n = 6, P < 0.001), 2 to 22.87 5.67%
(from 17.05 2.36 to 4.15 1.36 mN, n = 4, P <
0.001), 3 to 2.80 1.24% (from 12.32 1.34 to 2.80
1.24 mN, n = 7, P < 0.001), 4 to 23.93 4.11% (from
11.06 1.27 to 3.17 0.85 mN, n = 6, P < 0.001), 5 to
11.99 9.27% (from 17.25 2.45 to 0.84 1.18 mN,
n = 5, P < 0.001), 6 to 51.13 0.69% (from 13.58
3.05 to 6.96 1.56 mN, n = 5, P < 0.01), 7 to 9.00
5.22% (from 15.26 2.80 to 3.14 1.32 mN, n = 4,
P < 0.001), 8 to 30.55 1.20% (from 8.56 0.61 to
2.61 0.19 mN, n = 5, P < 0.001), 9 to 1.66 5.86%
(from 16.26 2.49 to 0.79 1.16 mN, n = 5, P < 0.001),
and 10 to 25.66 5.64% (from 14.72 2.40 to 3.04
0.98 mN, n = 6, P < 0.001). Resveratrol showed a reduc-
tion of fc to 35.23 2.00% (n = 5, P < 0.001).6 The
decrease in percent for 1, 6, 7, 8, and 9 is shown inFigure
1 and for 2, 3, 4, 5, and 10 is shown inFigure 2. The eval-
uation of all the synthesized compounds is listed in Table
2 (IC50 values and potency index related to resveratrol).
-50
-75
-100
0
1
3
10
30
100
concentration (µM)
Figure 1. Concentration-dependent effect of 1 (d), 6 (j), 7 (.), 8 (ꢀ),
and 9 ( ) on terminal ilea contracted by 60 mM KCl. The decrease in
percent of contraction force is semi logarithmically plotted on the
ordinate against the concentration of the compounds on the abscissa.
Symbols represent the arithmetic means SEM from
experiments.
4 to 6
sented compounds on terminal ilea. All compounds (1
to 10) produced a significant spasmolytic effect in a con-
centration-dependent manner without a significant ac-
tion on vascular smooth muscle and heart muscle
preparations. The most potent compound was com-
pound 3 (IC50 of 3.25 lM), which was 7.39 times more
potent than resveratrol, followed by 5 with an IC50 of
7.5 lM and 9 with an IC50 of 8.50 lM, which were
3.20 and 2.82 times more active compared to resveratrol.
Compounds 1, 2, 4, 7, 8, and 10 were less effective. Com-
pound 6 showed the weakest spasmolytic effect (IC50
over 100 lM). Due to the presented pharmacological re-
sults we are able to do some initial structure–activity
relationships. Two fluoro substituents in both meta posi-
tions of the acidic side of the amide linker demonstrated
a good dilating activity. In particular, compound 3 was
Spasmolytic drugs have severe side-effects such as QT-
prolongation and cardiac arrhythmias.7–10 Our results
demonstrate a selective spasmolytic activity of the pre-

5976
G. Brunhofer et al. / Bioorg. Med. Chem. 16 (2008) 5974–5981
0
-25
The results from this study demonstrate that new benzani-
lide derivatives presented herein act concentration depen-
dently as selective spasmolytic agents. There are different
mechanisms that are responsible for this effect. To identify
a possible involvement of cholinergic, serotoninergic and
histaminergic mechanisms we studied the effect of the sub-
stances on contractions caused by acetylcholine, seroto-
nin, and histamine, but there were no significant
inhibitory responses observable. Since it has been demon-
strated that an NO synthase is present in the mesenteric
plexus we investigated a possible involvement of NO.
However, inhibition of the NO synthase by nitro-L-argi-
nine did not show any change of the spasmolytic activity
of the substances (data not shown).11,12 Therefore, we as-
sume that large conductance calcium-activated potassium
channels known as BK channels or maxi-K which are ex-
pressed in various tissues are responsible for the selective
spasmolytic activity of the presented compounds. BK acti-
vating substances modulate muscular and neuronal hyper-
excitability and although maxi-K channels are even found
in the vascular system it is described that small molecules
are able to affect selectively BK channels present in the
intestinal tract.13–15 Hence, further experiments need to
be done to clarify the exact mechanism of action underly-
ing the selective spasmolytic activity of these highly inter-
esting compounds.
-50
-75
-100
0
1
3
10
30
100
Concentration (µM)
Figure 2. Concentration-dependent effect of 2 (s), 3 (h), 4 (,), 5 (e),
and 10 ( ) on terminal ilea contracted by 60 mM KCl. The decrease in
percent of contraction force is semi logarithmically plotted on the
ordinate against the concentration of the compounds on the abscissa.
Symbols represent the arithmetic means SEM from
experiments.
4 to 6
Table 2. IC50-values for the spasmolytic activity of the synthesized
benzanilide derivatives 1–10 (lM) and their potency index related to
resveratrol
Compound
IC50 (lM)
Potency index
4. Structure–activity relationship
1
42.00
27.50
3.25
0.57
0.87
7.39
0.80
3.20
>0.24
0.92
1.14
2.82
0.74
1
2
3
In order to better understand the spasmolytic activity of
compounds 1 to 10 we tried to correlate the IC50 value
to energetic, electrostatic, and structural parameters de-
rived from ab initio calculations.
4
30.00
7.50
5
6
>100
26.00
21.00
8.50
7
8
Thioamide analogs of the compounds 1 to 10 have also
been synthesized (results to be published). These show a
significant increase in spasmolytic activity and also exhi-
bit, due to the larger van der Waals radius of the sulfur
atom, a larger dihedral angle of the two benzene rings.
9
10
32.50
24.00
Resveratrol
the most potent substance in this series bearing a thio-
methyl substituent in para position on the anilino moi-
ety (IC50 3.25 lM). Exchange of the thiomethyl
substituent of the basic side of compound 3 against
3,4,5-trimethoxy (compound 2), 4-dimethylamino (com-
pound 4), 2,4,6-trifluoro (compound 7), or 4-fluoro sub-
stituents (compound 8) resulted in a decrease of the
spasmolytic activity (IC50 21.00–30.00 lM). But on the
other hand, 2,4-difluoro substituents at the aniline moi-
ety and two fluoro substituents in both m-positions of
the acidic side of the amide linker (compound 6) caused
a detrimental impact on the spasmolytic activity
(IC50 > 100 lM). Moreover, a trifluoromethoxy group
as well as a fluoro atom in the para position of the acidic
side together with dimethoxy or difluoro substituents in
both meta positions of the aniline moiety (compounds 5
and 9) showed also a good spasmolytic efficacy (IC50
7.50 and 8.50 lM). 2,4,6-Trifluoro substituents at the
basic side of the amide linker demonstrated a moder-
ately spasmolytic activity, which is not strictly depend-
ing on the substituents of the acidic side (compound 7:
3,5-difluoro, IC50 26.00 lM; compound 10: 3,5-dinitro,
IC50 32.50 lM; compound 1: 4-fluoro, IC50 42.00 lM).
We therefore started from the assumption, that the spas-
molytic action of compounds 1 to 10 is related to the con-
formation of the two benzene rings and the conformation
of the benzamide C–N bond. We tried to model the ener-
getic barrier of rotation around the C–N bond. The
amide bond was held fixed at intervals of 5° (compounds
1, 2, 3, and 4), respectively, 10° (compounds 5, 6, 7, 8, 9,
and 10) and let the rest of the molecule relax. The energy
of the optimized structures was sampled. The graphical
representation of the results can be seen in Figure 3.
We used the energies of the corresponding three energy
maxima and minima, the angles at these maxima and
minima and the Mulliken charges of the amide nitrogen,
the carbonyl carbon and the sum of the Mulliken
charges of the two benzene rings as independent vari-
ables of the subsequent regression analysis. Table 3
shows the resulting dihedral angles of the C–N bond
at the calculated respective minima and maxima.
Because of the large number of independent variables a
backward elimination algorithm was used including all

G. Brunhofer et al. / Bioorg. Med. Chem. 16 (2008) 5974–5981
5977
1
2
20
18
16
14
12
10
8
18
16
14
12
10
8
6
4
2
0
dE (kcal/mol)
dE (kcal/mol)
6
4
2
0
0
0
0
0
100
100
100
100
200
300
300
300
300
400
400
400
400
0
0
0
0
100
100
100
100
200
300
300
300
300
400
400
400
400
PHI
PHI
3
4
18
16
14
12
10
8
20
18
16
14
12
10
8
dE (kcal/mol)
dE (kcal/mol)
dE (kcal/mol)
dE (kcal/mol)
dE (kcal/mol)
dE (kcal/mol)
6
6
4
4
2
2
0
0
200
200
PHI
PHI
5
6
16
14
12
10
8
6
4
2
0
20
18
16
14
12
10
8
6
4
2
0
200
200
PHI
PHI
7
8
18
16
14
12
10
8
18
16
14
12
10
8
6
6
4
4
2
2
0
0
200
200
PHI
PHI
9
10
16
14
12
10
8
6
4
2
14
12
10
8
dE (kcal/mol)
6
dE (kcal/mol)
4
2
0
0
0
100
200
300
400
0
100
200
300
400
PHI
PHI
Figure 3. Potential energies of compounds 1 to 10 obtained through rotation of the C–N bond. The amide bond was held fix at intervals of 5°
(compounds 1, 2, 3, and 4), respectively, 10° (compounds 5, 6, 7, 8, 9, and 10).
independent variables and excluding variables stepwise
starting with the variable with the smallest partial corre-
lation and excluding variables with a F-value =0.100.
Three variables were identified as significant. The final
model was derived using the significant variables only
(Table 4). In order to get an impression of the predictiv-
ity of the model, we performed a leave-one-out cross-
validation of the final model. The q2 value of 0.60 indi-
cates a moderate predictivity of the model due to the
small number n of observations.
The plot of the calculated values derived from the
final model versus the experimental values is shown in
Figure 4. Figure 5 shows the plot of the calculated values

5978
G. Brunhofer et al. / Bioorg. Med. Chem. 16 (2008) 5974–5981
Table 3. AMIN1, AMIN2, AMIN3, AMAX1, AMAX2, and AMAX3 are the angles at minimum 1, 2, and 3 and at maximum 1, 2, and 3,
respectively
Compound
LNINVBIO
AMIN1
AMIN2
AMIN3
AMAX1
AMAX2
AMAX3
1
2
ꢀ3.95
ꢀ3.31
ꢀ1.18
ꢀ3.40
ꢀ2.01
ꢀ4.61
ꢀ3.26
ꢀ3.04
ꢀ2.14
ꢀ3.33
42.74000
31.96000
32.07000
27.45000
36.92000
37.29000
37.80000
32.27000
36.55000
37.03000
177.74001
176.96001
177.07001
177.45000
176.92000
177.28999
177.80000
177.27000
176.55000
177.03000
337.73999
341.95999
342.07001
342.45001
336.92001
337.29001
337.79999
337.26999
336.54999
337.03000
32.74000
26.96000
27.07000
22.45000
26.92000
27.29000
27.80000
27.27000
26.55000
27.03000
102.74000
101.96000
102.07000
112.45000
96.92000
262.73999
266.95999
262.07001
267.45001
256.92001
267.29001
257.79999
257.26999
256.54999
257.03000
3
4
5
6
107.29000
107.80000
107.27000
96.55000
7
8
9
10
107.03000
LNINVBIO is the natural logarithm of the inverse biological activity.
Table 4. Final model using three significant variables
showed a 7.39-fold higher activity than the comparing
compound resveratrol. Further studies investigating
the chronotropic, inotropic, or vasodilating activity of
the compounds proved the selectivity of the presented
compounds which would be a great asset compared to
other anti-diarrhoeal agents. We hypothesize that the
spasmolytic activity of the compounds is, at least partly,
due to the activation of BK-channels. Moreover, we
could show a correlation between the spasmolytic activ-
ity of compounds 1 to 10 and the benzamide C–N angle.
Due to the small amount of compounds under investiga-
tion the predictivity of the model is rather low.
Significant variables only
n = 10; r2 = 0.868; rc2v ¼ 0:601
Coefficient
t
SD
p
Intercept
AMAX2
AMAX3
AMIN3
ꢀ72.849
0.385
ꢀ3.293
4.648
22.19
0.083
0.032
0.054
0.017
0.004
0.028
0.005
ꢀ0.092
ꢀ0.195
ꢀ2.893
ꢀ4.365
The q2 value of 0.60 indicates a moderate predictivity of the model due
to the small number n of observations.
derived from the cross-validated final model versus the
experimental values.
6. Experimental
6.1. Chemistry
5. Conclusion
6.1.1. General experimental methods. All chemicals ob-
tained from commercial suppliers were used as received
and were of analytical grade. Melting points were deter-
We may conclude that this study demonstrates the syn-
thesis of new compounds possessing a relaxant effect on
guinea pig ileum and that especially compound 3
Figure 4. The calculated values predicted by the three component model are plotted against the experimental values.

G. Brunhofer et al. / Bioorg. Med. Chem. 16 (2008) 5974–5981
5979
Figure 5. The calculated values predicted by the cross-validated three component model (leave-one-out cross-validation) are plotted against the
experimental values.
mined on a Kofler hot stage apparatus and are uncor-
162.1 (JC,F = 247.0 Hz), 152.8, 138.3 (t, JC,F = 8.1 Hz),
134.7, 111.0 (JC,F = 26.0 Hz), 107.1, 98.1, 60.1, 55.7;
MS m/z 323 [M+, 27%], 141 [100%].
1
rected. The H- and 13C NMR spectra were recorded
on a Bruker Avance DPx200 (200 and 50 MHz). Chem-
ical shifts are reported in d units (ppm) relative to Me4Si
line as internal standard and J values are reported in
Hertz. Mass spectra were obtained by a Hewlett Pack-
ard (GC: 5890; MS: 5970) spectrometer. Solutions in or-
ganic solvents were dried over anhydrous sodium
sulfate.
6.1.2.3. N-(4-Methylthiophenyl)-3,5-difluorobenzamide
(3). Yield: 2.57 (92%), mp 175 °;. 1H NMR (d6-DMSO):
d 10.36 (s, 1H), 7.77–7.61 (m, 4H), 7.58–7.45 (m, 1H),
7.33–7.21 (m, 2H), 2.47 (s, 3H); 13C NMR (d6-DMSO):
d 162.7, 162.3 (JC,F = 247.0 Hz), 162.1 (JC,F = 247.0
Hz), 138.3, 136.0, 133.0, 126.7, 121.1, 111.1 (JC,F = 26.0
Hz), 107.0 (t, JC,F = 26.0 Hz), 15.3; MS m/z 279 [M+,
55%], 141 [100%].
6.1.2. General procedure for the synthesis of the benza-
nilide derivatives. To a solution of 10 mmol appropriate
benzoylchloride in 10 ml anhydrous dioxane was quickly
added a solution of 10 mmol substituted aniline deriva-
tive also dissolved in 10 ml anhydrous dioxane. The mix-
ture was shaken immediately. The reaction was carried
out in a 100 ml extraction funnel at room temperature.
The reaction mixture was allowed to stand for a period
of 10 min and then it was purged into ice water. The
formed solid was filtered and purified by recrystallization.
6.1.2.4. N-(4-Dimethylaminophenyl)-3,5-difluoroben-
1
zamide (4). Yield: 1.38 g (50%), mp 161 °C; H NMR
(CDCl3): d 7.85 (br s, 1H), 7.46–7.30 (m, 4H), 7.01–6.86
(m, 1H), 6.68 (JA,B = 8.0 Hz, 2H), 2.93 (s, 6H); 13C
NMR (CDCl3):
d Amid-C not detectable, 163.0
(JC,F = 251.6 Hz), 162.8 (JC,F = 251.6 Hz), 148.4, 126.7,
122.4, 112.8, 110.3 (d, JC,F = 26.3 Hz), 106.7 (t,
JC,F = 25.2 Hz), 40.7; MS m/z 276 [M+, 35%], 135 [100%].
6.1.2.1. N-(2,4,6-Trifluorophenyl)-4-fluorobenzamide
(1). Yield: 2.28 g (85%), mp 150 °C; 1H NMR (d6-
DMSO): d 10.17 (s, 1H), 8.13–7.97 (m, 2H), 7.49–7.25
(m, 4H); 13C NMR (d6-DMSO): d 164.5, 164.4
(JC,F = 247.1 Hz), 158.3 (JC,F = 249.6 Hz), 155.5–155.8
(m), 130.5 (JC,F = 9.5 Hz), 129.3 (JC,F = 2.9 Hz), 115.6
(JC,F = 22.5 Hz), 111.6–111.2 (m), 101.5–100.4 (m); MS
m/z 269 [M+, 19%], 123 [100%].
6.1.2.5. N-(3,5-Dimethoxyphenyl)-(4-trifluorometh-
oxy)benzamide (5). Yield: 2.34 g (68%), mp 111–
113 °C; H NMR (CDCl3): d 7.95–7.82 (m, 3H), 7.29
(d, JA,B = 8.0 Hz, 2H), 6.87 (d, J = 2.3 Hz, 2H), 6.28
(t, J = 2.3 Hz, 1H), 3.79 (s, 6H); 13C NMR (CDCl3): d
164.5, 161.1, 139.3, 133.3, 128.9, 120.7 (JC,F = 1.1 Hz),
98.4, 97.1, 55.3; MS m/z 341 [M+, 29%], 189 [100%].
1
6.1.2.2. N-(3,4,5-Trimethoxyphenyl)-3,5-difluoroben-
6.1.2.6. N-(2,4-Difluorophenyl)-3,5-difluorobenzamide
1
1
zamide (2). Yield: 2.85 g (88%), mp 186 °C; H NMR
(6). Yield: 2.05 g (76%), mp 143 °C; H NMR (CDCl3):
(d6-DMSO): d 10.25 (s, 1H,), 7.78–7.61 (m, 2H), 7.60–
7.45 (m, 1H), 7.21 (s, 2H), 3.78 (s, 6H), 3.65 (s, 3H);
13C NMR (d6-DMSO): d 162.5, 162.3 (JC,F = 247.0 Hz),
d 8.65 (s, 1H), 8.16–7.97 (m, 1H), 7.53–7.39 (m, 2H),
7.07–6.84 (m, 3H); 13C NMR (CDCl3): d 162.9 (JC,F
250.6 Hz), 162.7 (JC,F = 250.6 Hz), 161.7, 124.9–124.7
=

5980
G. Brunhofer et al. / Bioorg. Med. Chem. 16 (2008) 5974–5981
(m), 111.4–110.4 (m), 107.2 (t, JC,F = 25.2 Hz), 104.0
and 103.5 (dd, JC,F = 26.5 Hz and 23.4 Hz); MS m/z
269 [M+, 15%], 141 [100%].
were stored at room temperature in a Krebs–Henseleit
solution with the following composition (in mM) NaCl
114.9, KCl 4.73, CaCl2 3.2, MgSO4 1.18, NaHCO3,
24.9, KH2PO4 1.18, glucose 10; pH 7.2–7.4. Experiments
were performed at a temperature of 35 1 °C. The bath-
ing solutions were continuously bubbled by a mixture of
95% O2 and 5% CO2 to guarantee sufficient oxygen sup-
ply and appropriate pH as well as circulation of nutrient
solution with and without test substance.
6.1.2.7. N-(2,4,6-Trifluorophenyl)-3,5-difluorobenza-
mide (7). Yield: 2.10 g (74%), mp 171 °C; 1H NMR
(CDCl3): d 9.12 (s, 1H), 7.71–7.52 (m, 2H), 7.08–6.93
(m, 1H), 6.91–6.64 (m, 2H); 13C NMR (CDCl3): d
162.6 (JC,F = 249.8 Hz), 160.3, 111.4 and 111.1 (dd,
JC,F = 16.3 Hz and 10.2 Hz), 107.2 (t, JC,F = 25.2 Hz),
100.4 (dt, JC,F = 25.6 Hz and 3.3 Hz); MS m/z 287
[M+, 8%], 141 [100%].
Isometric contraction force of electrically stimulated
papillary muscles and spontaneous activity in right atria
was measured by the method described by Reiter.16
A
6.1.2.8. N-(4-Fluorophenyl)-3,5-difluorobenzamide (8).
force transducer and amplifier (TransbridgeTM, 4-Chan-
nel Transducer Amplifier, World Precision Instruments,
Sarasota, FL, USA) were used for measurements of iso-
metric contractions. Resting tension of 3.92 mN for
papillary muscles and 10.37 mN for right atria was kept
constant throughout the experiments. Papillary muscles
were electrically driven with an Anapulse Stimulator
Model 301-T and an Isolation Unit Model 305-1
(WPI, Hamden, CT, USA) at a frequency of 1 Hz and
pulse duration of 3 ms. Amplitude of stimulation pulse
was adjusted 10% above threshold level. After a control
period of 30 minutes the compounds 1–10 were added to
the bathing solution cumulatively, until a steady state
was reached. The aorta and the ileum were dissected
as well. The aorta was stored at room temperature in
gassed (95% O2 and 5% CO2), modified Krebs–Henseleit
solution with the following composition (in mM): NaCl
118.0, KCl 4.7, CaCl2 1.8, MgSO4 5.8, KH2PO4 1.4,
NaHCO3 11.9, and glucose 10. The aorta was cleaned
of loosely adhering fat and connective tissue and cut
into rings of 2 mm width. Aortic rings were precontract-
ed with 90 mM KCl.
1
Yield: 2.33 g (93%), mp 157 °C; H NMR (CDCl3): d
9.52 (s, 1H), 7.83–7.63 (m, 2H), 7.62–7.41 (m, 2H),
7.11–6.89 (m, 3H); 13C NMR (CDCl3): d 163.4, 162.7
(JC,F = 250.1 Hz), 162.5 (JC,F = 250.1 Hz), 159.3 (JC,F
=
243.8 Hz), 138.3 (JC,F = 8.1 Hz), 134.1 (JC,F = 2.7 Hz),
122.5 (JC,F = 8.1 Hz), 115.2 (JC,F = 22.6 Hz), 110.8
(JC,F = 10.0 Hz), 106.6 (t, JC,F = 25.2 Hz); MS m/z 251
[M+, 32%], 141 [100%].
6.1.2.9. N-(3,5-Difluorophenyl)-4-fluorobenzamide (9).
1
Yield: 2.25 g (90%), mp 155 °C; H NMR (CDCl3): d
7.95 (s, 1H), 7.90–7.80 (m, 2H), 7.30–7.09 (m, 4H),
6.65–6.54 (tt, JH,H = 8.8 Hz, JH,F = 2.3 Hz, 1H); 13C
NMR (CDCl3): d 165.1 (JC,F = 253.6 Hz), 164.7, 163.3
(JC,F = 247.0 Hz), 163.0 (JC,F = 247.0 Hz), 140.4
(JC,F = 13.4 Hz), 130.4, 129.5 (JC,F = 9.2 Hz), 116.0
(JC,F = 22.2 Hz), 103.2 (JC,F = 29.5 Hz), 99.9 (t,
JC,F = 25.7 Hz); MS m/z 251 [M+, 15%], 123 [100%].
6.1.2.10. N-(2,4,6-Trifluorophenyl)-3,5-dinitrobenzam-
1
ide (10). Yield: 2.39 g (70%), mp 232–236 °C; H NMR
(CDCl3): d 10.94 (s, 1H), 9.12 (s, 2H), 9.11–8.97 (m,
1H), 7.56–7.41 (m, 2H); 13C NMR (CDCl3): d 161.6,
148.4, 135.2, 128.1, 121.7, 101.2; MS m/z 341 [M+,
21%], 195 [100%].
The terminal portion of the ileum was removed and the
10 cm nearest to the caecum was discarded. The intes-
tine was placed in a nutrient solution. The intestine
was cleaned by flushing with nutrient solution and cut
into pieces of 1 to 2 cm length. The experiments were
performed at a temperature of 37 1°C. The smooth
muscle preparations were placed in a continuously oxy-
genated (95% O2 and 5% CO2) bath of 28 ml nutrient
solution with the following composition (in mM): NaCl
114.9, KCl 4.73, CaCl2 3.2, MgSO4 1.18, NaHCO3 24.9,
KH2PO4 1.18, glucose 10; pH 7.2–7.4. One end was con-
nected to a tissue holder and the other to a force trans-
ducer (TransbridgeTM, 4-Channel Transducer Amplifier,
World Precision Instruments, Sarasota, FL, USA).
The preparations were precontracted by 60 mM KCl
and a resting tension of 4.9 mN was kept constant
throughout the experiments. After a control period of
45 min the different concentrations of the compounds
were added to the bathing solution cumulatively, until
a steady state was reached.
6.1.3. Purity of compounds 1–10. The purity of the com-
pounds 1–10 was confirmed by elemental analysis and
was within 0.4% of the theoretical values calculated
for the structural formulas.
6.2. Pharmacological studies
Male and female guinea pigs (Him DH) (340–480 g)
were obtained from the Department of Laboratory
Zoology and Genetics (Medical University, Himberg,
Austria) and housed in an air-conditioned room with a
12-h photo period at a temperature of 22–24 °C and at
a relative humidity of 50–60%. The sex of the experi-
mental animal has no significant impact on the results
of these studies and therefore, both sexes were used.
The guinea pigs were killed with a blow on the neck.
After excision of the heart, papillary muscles were dis-
sected from the right ventricle for contractility measure-
ments. Purkinje fibers were carefully removed to prevent
spontaneous activity. Only muscles with a diameter of
less than 0.87 mm were used in order to have a sufficient
oxygen supply. Chronotropic activity was tested on gui-
nea pig isolated right atria. The isolated preparations
Signals were recorded with a chart recorder (BD 112
Dual Channel, Kipp & Zonen) and evaluated. For sta-
tistical analyses the arithmetic means and standard error
of the mean (SEM) of n experiments were calculated.
Statistical significance of the results was evaluated by
the Student’s t-test for paired observations.

G. Brunhofer et al. / Bioorg. Med. Chem. 16 (2008) 5974–5981
5981
´
3. Hernandez-Romero, Y.; Rojas, J. I.; Castillo, R.; Rojas,
A.; Mata, R. J. Nat. Prod. 2004, 67, 160.
Due to insolubility of the test compounds in aqueous
nutrient solution, stock solutions of the compounds 1–
10 were dissolved in dimethylsulfoxide (DMSO) every
day and were further diluted with modified Krebs–
Henseleit solution to the required concentrations. To ex-
clude the DMSO effect, a series of experiments with
DMSO only were performed at the same experimental
conditions. The DMSO effect was subtracted from the
results of the compounds.
4. Jang, M.; Cai, L.; Udeani, G. O.; Slowing, K. V.; Thomas,
C. F.; Beecher, C. W.; Fong, H. H.; Farnswoth, N. R.;
Kinghorn, A. D.; Mehta, R. G.; Moon, R. C.; Pezzuto, J.
M. Science 1997, 275, 218.
5. Navarette-Vazquez, G.; Moreno-Diaz, H.; Aguierre-Cre-
spo, F.; Leon-Rivera, I.; Villalobos-Molina, R.; Munoz-
Muniz, O.; Estrada–Soto, S. Bioorg. Med. Chem. Lett.
2006, 16, 4169.
6. Handler, N.; Brunhofer, G.; Studenik, C.; Leisser, K.;
Jaeger, W.; Parth, S.; Erker, T. Bioorg. Med. Chem. 2007,
15, 6109.
6.3. Computational methods
Molecular modeling was done using program MOE17 on
a Sun Ultra 40 workstation running Linux. Ab initio
7. Meyers, N. L.; Rickling, R. L. J. Int. Med. Res. 2007, 35,
848.
8. Toga, T.; Kohmura, Y.; Kawatsu, R. J. Pharmacol. Sci.
2007, 105, 207.
9. Mohammad, S.; Zhou, Z.; Gong, Q.; January, C. T. Am.
J. Physiol. 1997, 273, H2534.
18
calculations were done with GAMESS
machine and operating system.
on the same
Structures were derived using MOE and the MMFF94x
force field and a stochastic search algorithm as imple-
mented in MOE. The structure with the lowest energy
was taken as the starting point for the subsequent grid
search algorithm in GAMESS.
10. Paakkari, I. Toxicol. Lett. 2002, 127, 279.
11. Li, C. G.; Rand, M. J. Eur. J. Pharmacol. 1990, 191, 303.
12. Konturek, S. K.; Konturek, P. C. Digestion 1995, 56, 1.
13. Calderone, V. Curr. Med. Chem. 2002, 9, 1385.
14. Biagi, G.; Giorgi, I.; Livi, O.; Nardi, A.; Calderone, V.;
Martelli, A.; Martinotti, E.; LeRoy Salerni, O. Eur. J.
Med. Chem. 2004, 39, 491.
15. Sivarao, D. V.; Newberry, K.; Langdon, S.; Lee, A. V.;
Hewawasam, P.; Plym, M. J.; Signor, L.; Myers, R.;
Lodge, N. J. J. Pharmacol. Exp. Ther. 2005, 313, 840.
16. Reiter, M. Arzneim.-Forsch. 1967, 17, 1249.
17. MOE 2007.09 (Molecular Operating Environment). 2007,
Chemical Computing Group: 1010 Sherbrooke Street
West, Suite 910, Montreal Que., Canada H3A 2R7.
18. Schmidt, M. W.; Baldridge, K. K.; Boatz, J. A.; Elbert, S.
T.; Gordon, M. S.; Jensen, J. H.; Koseki, S.; Matsunaga,
N.; Nguyen, K. A.; Su, S.; Windus, T. L.; Dupius, M.;
Montgomery, J. A., Jr. J. Comput. Chem. 1993, 14, 1347.
19. SPSS, version 14, SPSS: Chicago IL, USA.
Statistical calculations were carried out with the multi-
ple linear regression module implemented in SPSS
14.0.19 The regression models were estimated by their
correlation coefficients r2 and the cross-validated corre-
lation coefficients q2 (leave-one-out cross-validation).
References and notes
1. Mertz, H. R. Gastroenterol. Clin. North Am. 2003, 32, 463.
2. Estrada, S.; Rojas, A.; Mathison, Y.; Israel, A.; Mata, R.
Planta Med. 1999, 65, 109.