Synthesis, spasmolytic activity and structure-activity relationship study of a series of polypharmacological thiobenzanilides

Article abstract of DOI:10.1016/j.ejps.2010.10.005

Recently we presented a series of benzanilide derivatives with a selective spasmolytic effect on terminal ileum preparations of the guinea pig. In this report we demonstrate a further development of these compounds. The exchange of the amide oxygen agains

Full text of DOI:10.1016/j.ejps.2010.10.005

European Journal of Pharmaceutical Sciences 42 (2011) 37–44
Contents lists available at ScienceDirect
European Journal of Pharmaceutical Sciences
journal homepage: www.elsevier.com/locate/ejps
Synthesis, spasmolytic activity and structure–activity relationship study of a
series of polypharmacological thiobenzanilides
Gerda Brunhofera, Christian Studenikb, Gerhard F. Eckerc, Thomas Erkera,∗
a Department of Medicinal Chemistry, University of Vienna, Althanstraße 14, 1090 Vienna, Austria
b Department of Pharmacology and Toxicology, University of Vienna, Althanstraße 14, 1090 Vienna, Austria
c Department of Medicinal Chemistry, Pharmacoinformatics Research Group, University of Vienna, Althanstraße 14, 1090 Vienna, Austria
a r t i c l e i n f o
a b s t r a c t
Article history:
Recently we presented a series of benzanilide derivatives with a selective spasmolytic effect on terminal
ileum preparations of the guinea pig. In this report we demonstrate a further development of these com-
pounds. The exchange of the amide oxygen against a sulfur atom resulted in an up to 325 fold increase of
the antispasmodic activity of the thiobenzanilide (IC50 of 0.1 ␮M) compared to its benzanilide derivative.
Considering their mode of action the compounds interacted with several molecular targets, suggesting
that we identified a chemical identity able to modulate multiple targets simultaneously. Furthermore,
based on this data set, we present a structure–activity relationship study supporting the important role
of the sulfur atom.
Received 25 June 2010
Received in revised form
23 September 2010
Accepted 11 October 2010
Available online 20 October 2010
Keywords:
Thiobenzanilide derivatives
Spasmolytic activity
Polypharmacology
© 2010 Elsevier B.V. All rights reserved.
Structure–activity relationship
1. Introduction
18 shows a 325 fold increase in spasmolytic activity compared to
its oxygen analogue. Investigations of the mechanism of action
Benzanilide derivatives showed a selective spasmolytic activity,
as displayed in the guinea pig ileum test, which correlates with the
N–H angle of the amide bond (Brunhofer et al., 2008). This class of
compounds was derived from the stilbenoid scaffold which repre-
sents a highly active principle. Resveratrol, gigantol, lusianthridin
and batatasin III are stilbenoids with remarkable spasmolytic activ-
ity (Estrada et al., 1999; Hernandez-Romero et al., 2004; Handler
et al., 2007). The basis of gastrointestinal disorders, like the irritable
bowel syndrome (IBS) which is characterized by symptoms such
as gastrointestinal hypermotility and abdominal pain, is still not
understood. Therefore, the principle of polypharmacology or the
modulation of several molecular targets could be a beneficial strat-
egy in the treatment of IBS (Hopkins et al., 2006). The compounds
presented are an example for so-called “multi-target-directed lig-
ands”, a single chemical entity able to modulate several targets at
the same time (Bolognesi et al., 2009). Within this study we mod-
ified the benzanilide scaffold and exchanged the amide oxygen
against a sulfur atom. The so-obtained thiobenzanilide derivatives
exhibit a significantly higher spasmolytic activity compared to their
corresponding benzanilide derivatives. The most active compound
reveal that the spasmolytic activity is a result of the modulation
of different molecular targets. The following study presents a fur-
ther development of a new class of compounds with spasmolytic
activity that demonstrates the feasibility of accessing a chemical
space which intersects several molecular targets thus represent-
ing possible treatment options for multifactorial diseases such
as IBS.
2. Materials and methods
2.1. Chemistry
2.1.1. General methods
All chemicals obtained from commercial suppliers were used as
received and were of analytical grade. Melting points were deter-
mined on a Kofler hot stage apparatus and are uncorrected. The 1
H
and 13C NMR spectra were recorded on a Bruker Avance DPx200
(200 and 50 MHz). Chemical shifts are reported in ( units (ppm) rel-
ative to Me4Si line as internal standard and J values are reported in
Hertz. Mass spectra were obtained by a Hewlett Packard (MS: 5970)
spectrometer. Solutions in organic solvents were dried over anhy-
drous sodium sulphate. The 1H, 13C NMR and mass spectra of the
new synthesized compounds are in agreement with the assigned
structures. C,H,N analysis data were within 0.4% of the theoretical
∗
Corresponding author. Tel.: +43 1 4277 55003; fax: +43 1 4277 9551.
E-mail address: thomas.erker@univie.ac.at (T. Erker).
0928-0987/$ – see front matter © 2010 Elsevier B.V. All rights reserved.
doi:10.1016/j.ejps.2010.10.005

38
G. Brunhofer et al. / European Journal of Pharmaceutical Sciences 42 (2011) 37–44
values. Column chromatography was performed using silica gel 60,
70-230 mesh ASTM (Merck).
2.1.9. N-(4-Fluorophenyl)-3,5-difluorobenzothioamide (15)
(Erker et al., 2007)
Yield 1.15 g (86%), mp 136–139 ◦C. 1H NMR (CDCl3, 200 MHz):
ı 8.96 (s, 1H, NH), 7.79–7.58 (m, 2H, phenyl-H), 7.48–7.28 (m, 2H,
phenyl-H), 7.22–7.05 (m, 2H, phenyl-H), 7.04–6.83 (m, 1H, phenyl-
H). 13C NMR (CDCl3, 50 MHz): ı 183.5, 162.8 (JC,F = 250.6 Hz), 162.5
(JC,F = 250.9 Hz), 161.0 (JC,F = 248.0 Hz), 134.4 (JC,F = 3.3 Hz), 126.0
(JC,F = 8.4 Hz), 116.0 (JC,F = 23.0 Hz), 110.0 (JC,F = 26.7 Hz), 106.4 (t,
JC,F = 25.2 Hz). MS: m/z 267 (M+, 28%), 157 (100%).
2.1.2. General procedure for the synthesis of the benzanilide
derivatives 1–10
The synthesis and analytical data of the benzanilide derivatives
1–10 are described in a previous study (Brunhofer et al., 2008).
2.1.3. General procedure for the synthesis of the thiobenzanilide
derivatives 11–24 and 27–30
To a solution of 5 mmol of the corresponding benzanilide deriva-
tive in 10 ml anhydrous THF 4 mmol (1.6 g) Lawesson’s reagent
were added and heated to reflux till the reaction was completed
(monitored by TLC). The organic solvent was removed in vacuo and
the so-obtained crude product was purified by column chromatog-
raphy or recrystallization.
2.1.10. N-(4-Dimethylaminophenyl)-3,5-difluorobenzothioamide
(16) (Erker et al., 2007)
Yield 0.85 g (58%), mp 116–117 ◦C. 1H NMR (CDCl3, 200 MHz):
ı 8.90 (s, 1H, NH), 7.53 (A-part of AB-system, JA,B = 8.9 Hz, 2H,
phenyl-H), 7.40–7.25 (m, 2H, phenyl-H), 7.02–6.84 (m, 1H, phenyl-
H), 6.71 (B-part of AB-system, JA,B = 8.9 Hz, 2H, phenyl-H), 2.98 (s,
6H, N(CH3)2). 13C NMR (CDCl3, 50 MHz): ı 162.7 (JC,F = 250.2 Hz),
149.4, 127.6, 124.9, 111.9, 110.3–109.7 (m), 105.9 (t, JC,F = 25.2 Hz),
40.4. MS: m/z 292 (M+, 75%), 259 (100%).
2.1.4. General procedure for the synthesis of the acetothioamide
derivatives 25 and 26
In a dry three-necked flask 10 mmol of the appropriate pheny-
lacetic acid derivative and 1 ml thionyl chloride were heated
to reflux for 2 h. After the removal of the thionyl chloride
in vacuo 8 ml anhydrous 1,4-dioxane, 10 mmol (1.5 ml) 2,4-
dimethoxybenzylamine and 10 mmol (1.0 ml) triethylamine were
added. The so-obtained acetoamide was recrystallized in ethanol.
The synthesis of the corresponding acetothioamide derivative fol-
lowed the procedure described in Section 2.1.3.
2.1.11. N-(4-Methylthiophenyl)-4-fluorobenzothioamide (17)
(Erker et al., 2007)
Yield 0.78 g (56%), mp 148–152 ◦C. 1H NMR (CDCl3, 200 MHz):
ı 8.95 (s, broad, 1H, NH), 7.89–7.76 (m, 2H, phenyl-H), 7.75–7.53
(m, 2H, phenyl-H), 7.36–7.19 (m, 2H, phenyl-H), 7.18–6.96 (m, 2H,
phenyl-H), 2.49 (s, 3H, SCH3). 13C NMR (CDCl3, 50 MHz): ı 135.9,
129.0–128.8 (m), 126.7, 124.2, 115.5 (JC,F = 22.0 Hz), 15.7. MS: m/z
277 (M+, 25%), 139 (100%).
2.1.5. N-(4-Methylthiophenyl)-3,5-difluorobenzothioamide (11)
(Erker et al., 2007)
2.1.12. N-(2,4,6-Trifluorophenyl)-3,5-dinitrobenzothioamide
(18) (Erker et al., 2007)
Yield 0.58 g (39%), mp 142–145 ◦C. 1H NMR (CDCl3, 200 MHz):
ı 8.92 (s, 1H, NH), 7.71–7.65 (m, 2H, phenyl-H), 7.43–7.20 (m,
4H, phenyl-H), 7.05–6.84 (m, 1H, phenyl-H), 2.50 (s, 3H, SCH3).
13C NMR (CDCl3, 50 MHz): ı 137.9, 135.5, 126.7, 124.0, 110.0
(JC,F = 26.7 Hz), 106.3 (t, JC,F = 26.3 Hz), 15.7. MS: m/z 295 (M+, 46%),
156 (100%).
Yield 1.23 g (70%), mp 223–225 ◦C. 1H NMR (d6-DMSO,
200 MHz): ı 12.25 (s, 1H, NH), 9.16–8.96 (m, 3H, phenyl-H),
7.59–7.33 (m, 2H, phenyl-H). 13C NMR (d6-DMSO, 50 MHz):
1
2
ı
195.3, 163.7, 159.8 (q, JC,F = 29.1 Hz, JC,F = 64.0 Hz), 155.1
1
2
(q, JC,F = 29.1 Hz, JC,F = 64.0 Hz), 147.8, 140.8, 127.6, 120.9,
102.2–101.1 (m). MS: m/z 357 (M+, 28%), 69 (100%).
2.1.6. N-(2,3-Dimethylphenyl)-4-methoxybenzothioamide (12)
(Erker et al., 2007)
2.1.13. N-(3,4,5-Trimethoxyphenyl)-3,5-dinitrobenzothioamide
(19) (Erker et al., 2007)
Yield 0.92 g (68%), mp 150 ◦C. 1H NMR (CDCl3, 200 MHz): ı 8.67
(s, broad, NH), 7.97–7.86 (m, 2H, phenyl-H), 7.27–7.13 (m, 3H,
phenyl-H), 6.96–7.84 (m, 2H, phenyl-H), 3.86 (s, 3H, OCH3), 2.33
(s, 3H, CH3), 2.19 (s, 3H, CH3). 13C NMR (CDCl3, 200 MHz): ı 193.0,
138.1, 129.6, 128.7, 126.0, 124.7, 113.6, 55.5, 20.4, 14.2. MS: m/z
271 (M+, 11%), 256 (100%).
Yield 1.69 g (86%), mp 202–210 ◦C. 1H NMR (CDCl3, 200 MHz): ı
11.58 (s, broad, 1H, NH), 9.05 (s, 3H, phenyl-H), 7.25 (s, 2H, phenyl-
H), 3.87 (s, 9H, OCH3). 13C NMR (CDCl3, 50 MHz): ı 190.8, 152.8,
147.6, 145.3, 135.0, 127.5, 119.3, 101.0, 60.7, 56.0. MS: m/z 393
(M+, 25%), 139 (100%).
2.1.7. N-(3,4,5-Trimethoxyphenyl)-3,5-difluorobenzothioamide
(13) (Erker et al., 2007)
2.1.14. N-(2,6-Dimethylphenyl)-3,5-dimethoxybenzothioamide
(20) (Erker et al., 2007)
Yield 1.07 g (63%), mp 174–175 ◦C. 1H NMR (CDCl3, 200 MHz): ı
9.05 (s, 1H, NH), 7.41–7.25 (m, 2H, phenyl-H), 7.09 (s, 2H, phenyl-
H), 7.07–6.85 (m, 1H, phenyl-H), 3.84 (s, 9H, OCH3). 13C NMR
(CDCl3, 50 MHz): ı 153.2, 134.4, 109.9 (JC,F = 26.7 Hz), 106.2 (t,
JC,F = 26.3 Hz), 100.8, 60.9, 56.1. MS: m/z 339 (M+, 45%), 157 (100%).
Yield 1.29 g (86%), mp 114 ◦C. 1H NMR (CDCl3, 200 MHz): ı 8.65
(s, 1H, NH), 7.22–7.09 (m, 3H, phenyl-H), 7.04 (d, JH,H = 2.3 Hz, 2H,
phenyl-H), 6.62–6.53 (t, 1H, phenyl-H), 3.84 (s, 6H, OCH3), 2.28 (s,
6H, CH3). 13C NMR (CDCl3, 50 MHz): ı 199.0, 160.7, 143.6, 136.1,
135.5, 128.4, 128.4, 104.9, 103.1, 55.6, 18.0. MS: m/z 301 (M+, 13%),
286 (100%).
2.1.8. N-(3,5-Difluorophenyl)-4-fluorobenzothioamide (14)
(Erker et al., 2007)
Yield 0.84 g (63%), mp 145 ◦C. 1H NMR (CDCl3, 200 MHz): ı
8.93 (s, 1H, NH), 7.83–7.76 (m, 2H, phenyl-H), 7.40–7.37 (m,
2H, phenyl-H), 7.10 (t, JH,H = 8.6 Hz, 2H, phenyl-H), 6.97–6.68
(tt, JH,H = 8.6 Hz, JH,F = 2.3 Hz, 1H, phenyl-H). 13C NMR (CDCl3,
50 MHz): ı 186.9, 164.7 (JC,F = 253.6 Hz), 163.0 (JC,F = 248.5 Hz),
162.8 (JC,F = 248.5 Hz), 148.8, 140.8, 129.0 (JC,F = 8.8 Hz), 115.8
(JC,F = 21.8 Hz), 106.5 (JC,F = 29.5 Hz), 102.2 (t, JC,F = 25.5 Hz). MS: m/z
267 (M+, 19%), 139 (100%).
2.1.15.
N-(2,6-Dimethylphenyl)-3,4,5-trimethoxybenzothioamide (21)
(Erker et al., 2007)
Yield 1.23 g (74%), mp 209 ◦C. 1H NMR (CDCl3, 200 MHz): ı 8.64
(s, 1H, NH), 7.30–7.11 (m, 5H, phenyl-H), 3.94 (s, 6H, OCH3), 3.89
(s, 3H, OCH3), 2.31 (s, 6H, CH3). 13C NMR (CDCl3, 50 MHz): ı 198.7,
153.0, 137.1, 136.3, 135.5, 128.5, 128.4, 104.4, 60.9, 56.3, 18.1. MS:
m/z 331 (M+, 19%), 316 (100%).

G. Brunhofer et al. / European Journal of Pharmaceutical Sciences 42 (2011) 37–44
39
H), 7.38–7.10 (m, 3H, aromat. H), 2.50 (s, 3H, SCH3). 13C NMR (d6-
DMSO, 50 MHz): ı 186.6, 148.4, 136.7, 136.3, 135.0, 128.4, 126.0
(2C), 125.7 (2C), 125.2, 15.0. MS: m/z 265 (M+, 24%), 127 (100%).
2.1.16. N-(2,4,6-Trifluorophenyl)-3,5-difluorobenzothioamide
(22)
Yield 0.88 g (83%), mp 150 ◦C. 1H NMR (CDCl3, 200 MHz): ı
11.68 (s, 1H, NH), 7.77–7.30 (m, 5H, phenyl-H). 13C NMR (CDCl3,
50 MHz): ı 197.8, 164.6–155.2 (5C), 142.6 (JC,F = 9 Hz), 114.0–113.9
(1C), 111.2 (dd, JC,F = 10.2 Hz, JC,F = 6.2 Hz), 106.9 (t, JC,F = 26 Hz),
101.6 (dt, JC,F = 26 Hz, JC,F = 4 Hz). MS: m/z 303 (M+, 25%), 157 (100%).
2.1.23. N-(3,4,5-Trimethoxyphenyl)-thionicotinicamide (29)
(Erker et al., 2007)
Yield 0.29 g (19%), mp 190–191 ◦C. 1H NMR (d6-DMSO,
200 MHz): ı 11.93 (s, 1H, NH), 8.92 (d, JH,H = 1.9 Hz, 1H, aromat. H),
8.70–8.67 (m, 1H, aromat. H), 8.17–8.11 (m, 1H, aromat. H), 7.39
(s, 2H, aromat. H), 3.77 (s, 6H, OCH3), 3.69 (s, 3H, OCH3). 13C NMR
(d6-DMSO, 50 MHz): ı 193.9, 152.5 (2C), 151.2, 147.5, 138.7, 135.7,
135.6, 135.1, 123.1, 101.4 (2C), 60.2, 56.0 (2C). MS: m/z 304 (M+,
67%), 122 (100%).
2.1.17. N-Benzyl-3,4,5-trimethoxybenzothioamide (23) (Erker
et al., 2007)
Yield 1.40 g (72%), mp 122 ◦C. 1H NMR (CDCl3, 200 MHz): ı 7.81
(s, 1H, NH), 7.39 (s, 5H, phenyl-H), 6.95 (s, 2H, phenyl-H), 4.99 (d,
JH,H = 5.3 Hz, 2H, CH2), 3.85 (s, 6H, OCH3), 3.80 (s, 3H, OCH3). 13C
NMR (CDCl3, 50 MHz): ı 198.7, 152.8, 140.5, 137.4, 136.1, 128.9,
128.3, 128.1, 104.2, 60.7, 56.2, 51.0. MS: m/z 317 (M+, 75%), 91
(100%).
2.1.24. N-(4-Methylthiophenyl)-thionicotinicamide (30) (Erker
et al., 2007)
Yield 0.60 g (46%), mp 140–141 ◦C. 1H NMR (d6-DMSO,
200 MHz): ı 11.64 (s, 1H, NH), 8.62 (d, JH,H = 1.3 Hz, 1H, aromat. H),
8.40–8.34 (m, 1H, aromat. H), 7.85–7.81 (m, 1H, aromat. H), 7.50 (A-
part of AB-system, d, JA,B = 8.6 Hz, 2H, aromat. H), 7.20–7.14 (m, 1H,
aromat. H), 7.01 (B-part of AB-system, d, JA,B = 8.6 Hz, 2H, aromat.
H), 2.18 (s, 3H, SCH3). 13C NMR (d6-DMSO, 50 MHz): ı 194.2, 151.2,
147.6, 138.3, 136.8, 136.2, 135.1, 125.9 (2C), 124.4 (2C), 123.1, 14.8.
MS: m/z 260 (M+, 24%), 122 (100%).
2.1.18.
N-(2,4-Dimethoxybenzyl)-3,4,5-trimethoxybenzothioamide (24)
(Erker et al., 2007)
Yield 0.98 g (52%), mp 115 ◦C. 1H NMR (CDCl3, 200 MHz): ı 7.95
(s, broad, 1H, NH), 7.32 (d, JH,H = 8.7 Hz, 1H, phenyl-H), 6.95 (s, 2H,
phenyl-H), 6.51–6.45 (m, 2H, phenyl-H), 4.92 (d, JH,H = 5.2 Hz, 2H,
CH2), 3.87 (s, 6H, OCH3), 3.86 (s, 3H, OCH3), 3.84 (s, 3H, OCH3),
3.81 (s, 3H, OCH3). 13C NMR (CDCl3, 50 MHz): ı 197.5, 161.0, 158.7,
152.8, 140.4, 137.7, 131.5, 116.4, 104.2, 104.1, 98.7, 60.8, 56.2, 55.3,
46.7. MS: m/z 377 (M+, 23%), 151 (100%).
2.1.25. 1-Methyl N1-(2,4,6-trifluorophenyl)-3,5-dinitro-1-
benzenecarboximidothioate
(31)
Yield 0.82 g (44%), mp 113 ◦C. 1H NMR (CDCl3, 200 MHz): ı
9.31–8.22 (m, 3H, phenyl-H), 7.00–6.89 (m, 2H, phenyl-H), 2.70 (s,
3H, CH3). 13C NMR (CDCl3, 50 MHz): ı 174.1, 171.3, 150.1, 148.4,
127.1, 120.0, 101.1–100.1, 14.7. MS: m/z 371 (M+, 16%), 75 (100%).
2.1.19. N-(2,4-Dimethoxybenzyl)-2-(3,4-
dimethoxyphenyl)acetothioamide (25) (Erker et al.,
2007)
Yield 0.78 g (43%), mp 98 ◦C. 1H NMR (CDCl3, 200 MHz): ı 7.63 (s,
broad, 1H, NH), 7.25–7.12 (m, 1H, phenyl-H), 6.84 (d, JH,H = 8.0 Hz,
1H, phenyl-H), 6.80–6.65 (m, 2H, phenyl-H), 6.45–6.34 (m, 2H,
phenyl-H), 4.74 (d, JH,H = 5.4 Hz, 2H, CH2), 4.07 (s, 2H, CH2), 3.88 (s,
6H, OCH3), 3.80 (s, 3H, OCH3), 3.78 (s, 3H, OCH3), 3.58 (s, 3H, OCH3).
13C NMR (CDCl3, 50 MHz): ı 200.7, 160.8, 158.4, 149.3, 148.5, 131.0,
127.0, 122.0, 116.3, 112.4, 111.3, 103.8, 98.4, 55.9, 55.8, 55.3, 54.9,
52.6, 46.1. MS: m/z 361 (M+, 20%), 151 (100%).
2.2. Pharmacology
To evaluate the spasmolytic activity of the compounds isometric
contraction measurements were used according to the procedure
described by Brunhofer et al. (2008) and Galanski et al. (2005).
Due to insolubility of the test compounds in aqueous nutri-
ent solution, stock solutions of the compounds were dissolved in
dimethylsulfoxide (DMSO) every day and were further diluted with
modified Krebs–Henseleit solution to the required concentrations.
To exclude 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.
To elucidate a possible mode of action the effect of the com-
pounds was studied in presence of glibenclamide (ATP-dependent
potassium (KATP) channel inhibitor), acetylcholine, histamine and
phenylephrine in terminal ileum preparations of the guinea pig.
For this set of experiments the concentration of the thioamide
derivatives used represented approximately their IC50 values. To
investigate whether KATP-channels are involved, experimental con-
ditions were set up as described elsewhere (Brunhofer et al., 2008).
After the steady-state was reached either 30 ␮M or 100 ␮M gliben-
clamide, dissolved in DMSO, were added to the organ bath. After
45 min the appropriate IC50 concentration of the corresponding test
compound was added.
2.1.20.
N-(2,4-Dimethoxybenzyl)-2-(4-methoxyphenyl)acetothioamide
(26) (Erker et al., 2007)
Yield 0.61 g (37%), mp 71 ◦C. 1H NMR (CDCl3, 200 MHz): ı 7.61 (s,
1H, NH), 7.30–7.06 (m, 4H, phenyl-H), 6.93–6.83 (m, 2H, phenyl-H),
6.45–6.34 (m, 2H, phenyl-H), 4.73 (d, JH,H = 5.4 Hz, 2H, CH2), 4.06 (s,
2H, CH2), 3.80 (s, 3H, OCH3), 3.78 (s, 3H, OCH3), 3.59 (s, 3H, OCH3).
13C NMR (CDCl3, 50 MHz): ı 200.9, 160.8, 159.0, 158.4, 131.0, 130.8,
126.6, 116.3, 114.3, 103.8, 98.5, 55.3, 55.2, 55.0, 52.1, 46.2, 39.2. MS:
m/z 331 (M+, 31%), 151 (100%).
2.1.21. N-(3,4,5-Trimethoxyphenyl)-2-thiophenecarbothioamide
(27) (Erker et al., 2007)
Yield 0.90 g (58%), mp 105–107 ◦C. 1H NMR (d6-DMSO,
200 MHz): ı 11.44 (s, broad, 1H, NH), 7.94–7.79 (m, 2H, aromat.
H), 7.30–7.10 (m, 3H, aromat. H), 3.77 (s, 6H, OCH3), 3.68 (s, 3H,
OCH3). 13C NMR (d6-DMSO, 50 MHz): ı 186.4, 152.7 (2C), 148.6,
135.8, 135.5, 135.0, 128.4, 125.0, 102.7 (2C), 60.3, 56.2 (2C). MS:
m/z 309 (M+, 20%), 127 (100%).
For the investigation of the effect of the test compounds in
presence of acetylcholine, histamine and phenylephrine the same
experimental set up was used as described by Brunhofer et al., 2008.
After a resting period of 20 min different concentrations of phenyle-
phrine (0.1, 0.3, 1, 3, and 10 ␮M), acetylcholine or histamine (0.01,
0.03, 0.1, 0.3, 1, and 3 ␮M) were added to the organ bath. For the
control values, only the effect of the different concentrations of
2.1.22. N-(4-Methylthiophenyl)-2-thiophenecarbothioamide
(28) (Erker et al., 2007)
Yield 0.94 g (71%), mp 136–138 ◦C. 1H NMR (d6-DMSO,
200 MHz): ı 11.50 (s, broad, 1H, NH), 7.90–7.58 (m, 4H, aromat.

40
G. Brunhofer et al. / European Journal of Pharmaceutical Sciences 42 (2011) 37–44
Table 1
Structures of the synthesized benzanilide (1–10) and thiobenzanilide derivatives 11–30 as well as of compound 31 and their biological activity on terminal ileum preparations.
H3C
F
F
X
S
Ar
O2N
n
N
m
N
H
R
F
31
1-30
NO2
.
Compound
X
Ar
n
m
R
IC50 (␮M)
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
O
O
O
O
O
O
O
O
O
O
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
S
4-F-C6H4
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
0
1
1
1
1
0
0
0
0
2,4,6-F3
3,4,5-(OCH3)3
4-SCH3
4-N(CH3)2
3,5-(OCH3)2
2,4-F2
2,4,6-F3
4-F
2,4-F2
2,4,6-F3
52
3,5-F2-C6H3
3,5-F2-C6H3
3,5-F2-C6H3
4-OCF3-C6H4
3,5-F2-C6H3
3,5-F2-C6H3
3,5-F2-C6H3
4-F-C6H4
3,5-(NO2)2-C6H3
3,5-F2-C6H3
4-OCH3-C6H4
3,5-F2-C6H3
4-F-C6H4
3,5-F2-C6H3
3,5-F2-C6H3
4-F-C6H4
3,5-(NO2)2-C6H3
3,5-(NO2)2-C6H3
3,5-(OCH3)2-C6H3
27.5
3.25
30
7.5
100
26
21
8.5
28
3
11.4
6.3
5.5
4.7
8.1
5.5
0.1
12.5
11.3
23.5
2.2
12.9
6.4
8.4
6.6
2.1
2.7
57.7
21.5
28
4-SCH3
2,3-(CH3)2
3,4,5-(OCH3)3
3,5-F2
4-F
4-N(CH3)2
4-SCH3
2,4,6-F3
3,4,5-(OCH3)3
2,6-(CH3)2
2,6-(CH3)2
2,4,6-F3-C6H2
H
2,4-(OCH3)2
2,4-(OCH3)2
2,4-(OCH3)2
3,4,5-(OCH3)3
4-SCH3
3,4,5-(OCH3)3-C6H2
3,5-F2-C6H3
3,4,5-(OCH3)3-C6H2
3,4,5-(OCH3)3-C6H2
3,4-(OCH3)2-C6H3
4-OCH3-C6H4
2-Thienyl
2-Thienyl
3-Pyridinyl
3-Pyridinyl
S
S
S
3,4,5-(OCH3)3
4-SCH3
phenylephrine, acetylcholine and histamine were recorded imme-
diately for 30 s. Then, the solution was discharged and the ileum
preparation was washed two times with Krebs–Henseleit solution.
After 5 min the appropriate IC50 concentration of the test com-
pound was added for 45 min. Afterwards, to elucidate the possible
effect of the test compounds, acetylcholine, histamine or phenyle-
phrine induced contraction was measured in the presence of the
test compound.
3. Results and discussion
3.1. Chemistry
The thiobenzanilide derivatives were synthesized by the reac-
tion of the corresponding benzanilide derivatives with Lawesson’s
reagent. The reaction was carried out under anhydrous condi-
tions in tetrahydrofuran at 67 ◦C and afforded all compounds in
good yield. The corresponding benzanilide derivatives were syn-
thesized by using a standard chemical methodology. The reaction
between the appropriate benzoyl chloride and the correspond-
ing substituted aniline yielded the desired benzanilide derivative
(Brunhofer et al., 2008). To synthesize the acetothioamides 25 and
26 the appropriate phenylacetic acid derivative was first converted
into the corresponding acid chloride by using thionyl chloride
(Scheme 1). Table 1 gives an overview of the synthesized com-
pounds as well as their pharmacological activity (IC50 values) on
isolated terminal ileum preparations.
2.3. Structure–activity relationship
2.3.1. Free-Wilson analysis
For each unique substitution pattern on ring 1 (Ar, Table 1) and
ring 2 (R-Phenyl, Table 1) as well as for X, indicator variables were
created. Compounds 23–26 and 31 were deleted from the data set
as they did not fit the general encoding scheme. Subsequently, all
variables which are present in the data set only once were deleted.
This led to the removal of 8 indicator variables and five compounds
(5, 6, 12, 20, 21). The final data set was submitted to multiple linear
regression analysis in Excel 7.0. Stepwise backward selection in
order to retain statistically significant variables (P < 0.1) gave the
following final equation (Eq. (1)):
3.2. Pharmacology and structure–activity relationship
pIC50 = 4.51–0.63(4-F; ring 1) − 1.10 (3-pyridyl; ring 1)
+ 0.78 (X = S) + 0.51 (4-SCH3; ring 2)
The inotropic and chronotropic effects of the compounds were
studied on isolated heart muscle preparations of guinea pigs in a
concentration range between 0.1 and 100 ␮M. None of the thioben-
zanilide compounds showed a significant change in inotropy or
+ 0.89 (3, 5-diF; ring 2) + 0.49 (2, 4, 6-triF; ring 2)
n = 21, r2 = 0.66, F = 4.50
(1)

G. Brunhofer et al. / European Journal of Pharmaceutical Sciences 42 (2011) 37–44
41
O
O
(a)
Ar
Ar
H2N
m
+
n
N
H
m
n
Cl
R
R
1-10
(b)
S
SCH3
(c)
Ar
Ar
n
N
m
n
N
H
m
R
R
31
11-30
Scheme 1. General synthetic route for compounds 1–31. Conditions: (a) triethylamine, 1,4-dioxane, room temperature; (b) Lawesson’s reagent, tetrahydrofuran, reflux; (c)
NaH, CH3I, tetrahydrofuran, room temperature. m, n = 0, 1.
chronotropy up to a concentration of 100 ␮M (data not shown)
which is comparable to the benzanilide derivatives (Brunhofer
et al., 2008). The spasmolytic effect was tested on terminal ileum
preparations which are a very common and reliable method for
the identification of antispasmodic compounds (Gutierrez and
Solis, 2009; Shah et al., 2010). Contraction of the preparation
was induced chemically by 60 mM KCl solution. All compounds
reversed concentration-dependently the contraction evoked by
KCl: at the highest concentration studied (100 ␮M for compounds
1, 3, 5–12, 14, 17, 19–21, and 23–31, 30 ␮M for compounds 2,
4, 13, 15, 16, and 22, and 1 ␮M for compound 18) the ben-
zanilide derivatives 2 and 4 reduced contraction force (fc) by
53.44 4.98% and 50.22 9.08% compared to their sulphur ana-
logues 13 and 16, which showed a reduction of fc by 91.65 2.64%
and 89.10 3.99%, respectively. The next couples of benzanilide
and thiobenzanilide derivatives, compounds 7 and 22, and 8
and 15, show the same effect. The benzanilides reduced fc by
56.05 8.60% (7) and 64.74 3.79% (8) whereas the correspond-
ing thiobenzanilides reached a reduction of 96.75 2.70% (22) and
97.25 5.64% (15) at a concentration of 30 ␮M. The most obvious
difference in spasmolytic activity showed the compounds 10 and
18. Compound 18 showed a reduction of fc by 96.90 4.29% even at
a concentration of 1 ␮M whereas its benzanilide derivative reduced
fc only by 6.29 1.44% and at a concentration of 100 ␮M compound
10 reached only 64.34 5.64% reduction of fc. This direct compari-
son of benzanilide and thiobenzanilide derivatives having the same
substitution pattern on both aromatic rings showed that the sul-
phur analogue is about five times more potent than the oxygen
derivative. An exception is compound 18, the most potent anti-
spasmodic compound, which was 325 times more active than the
corresponding benzanilide derivative 10 (Fig. 1A–D). Elongation
of the thioamide linker by one or two methylene groups (com-
pounds 23–26) resulted in no change of the spasmolytic potency.
2,4-Dimethoxy substituents on the amine side of the molecules
showed a slightly higher activity (IC50 of compound 24 6.4 ␮M)
compared to an unsubstituted aromatic ring (IC50 of compound 23
12.9 ␮M).
possible mode of action for most of the compounds. Only the data
for compound 28 are shown as an example in this publication. We
did not show the data for the most or least potent compound but
chose compound 28 that is under the top five most potent com-
pounds. Fig. 2 shows exemplarily the effect of 2 ␮M (IC50-value) of
compound 28 on KATP channels, histamine receptors, acetylcholine
receptors and gralpha-adrenoceptors.
We were able to identify several mechanisms which are respon-
sible for the antispasmodic effect of the thiobenzanilide derivatives.
The activity of the herein presented compounds is the result of
an activation of gralpha-adrenoceptors as well as an inhibition
of histamine receptors. Moreover, we suppose, that there is also
an influence of Ca2+-channels (Godfraind et al., 1986). Activation
of KATP channels and acetylcholine receptors is not significantly
involved. The concept of polypharmacology is increasingly gain-
ing acceptance as a way to treat multifactorial diseases (Hopkins,
2008). It has already been successfully applied to Alzheimer’s dis-
ease, oncogenic mutations, some central nervous system disorders,
and could also be a method for developing effective antibacterial
drugs (Bolognesi et al., 2009; Roth et al., 2004; Apsel et al., 2008;
Rasko et al., 2008). Therefore, the compounds presented, which are
able to modulate multiple targets simultaneously, could be bene-
ficial, for example, in the treatment of IBS.
Furthermore, based on this biological data set, we were able
to perform preliminary structure–activity relationship studies. The
thiobenzanilide 18 with meta-dinitro substitution at ring 1 and
2,4,6-trifluoro substitution on ring 2 shows the highest biologi-
cal activity. Pair wise comparison of compounds revealed several
qualitative trends. Considering ring 1, di-nitro (18) is by a factor of
20 better than the respective di-fluoro derivative (22). However,
this is only valid for a trifluoro-substitution at ring 2. If ring 2 is
substituted with 3,4,5-trimethoxy, the order is reversed: 2-thienyl
(27) > difluoro (13) > dinitro (19) > 3-pyridyl (30). In contrast, the
observation that 2-thienyl is generally favoured over 3-pyridyl is
consistent for both 17/19 and 18/20. At ring 2 the most remarkable
effect is the 100 fold decrease of activity when exchanging 2,4,6-
trifluoro (18) by 3,4,5-trimethoxy (19). However, once more this is
only seen for a dinitro-substitution at ring 1. In case of 3,5-difluoro
at ring 1, there is almost no effect if the pattern at ring 2 changes (13
vs 22). When comparing thioanilides with anilides, no clear pattern
can be observed showing differences ranging from almost equal
activity (11 vs 3) via a moderate effect (13 > 2 and 16 > 24) to a more
than 200 fold decrease for anilides when compared to thioanilides
(18 ꢀ 10). One more compound 18 seems to be exceptional. To
shed more light on this in part rather complex SAR we performed
To elucidate the mode of action of these compounds sev-
eral experiments were performed targeting different molecular
mechanisms. First, the impact of KATP channels on the spas-
molytic effect of the compounds was studied using glibenclamide,
a selective KATP channel inhibitor. To get information about a pos-
sible involvement of histamine receptors, acetylcholine receptors
and gralpha-adrenoceptors, histamine, acetylcholine and phenyle-
phrine were used as agonists. We have done the studies of a

42
G. Brunhofer et al. / European Journal of Pharmaceutical Sciences 42 (2011) 37–44
A
25
0
B
25
IC50 benzanilide
PI =
= 4.4
IC50 thiobenzanilide
PI = 3.7
0
-25
OMe
OMe
OMe
-25
-50
-75
-100
O
CH
N
F
N
H
O
CH
F
2
N
H
F
-50
OMe
4
OMe
OMe
F
CH
N
S
F
-75
S
CH
N
H
F
N
H
13
F
-100
16
F
concentration (µM)
Concentration (µM)
C
D
25
0
25
0
PI = 4.5
PI = 11.8
-25
-50
-75
-100
-25
-50
-75
-100
F
F
F
F
O
O
F
F
N
H
N
H
F
F
7
8
F
F
F
F
F
F
S
S
F
F
N
H
N
H
22
15
Concentration (µM)
Concentration (µM)
E
25
0
PI = 325
F
F
O
-25
-50
-75
-100
O N
N
H
F
10
NO
NO
F
F
S
O N
N
H
F
18
Concentration (µM)
Fig. 1. Concentration-dependent effect on terminal ileum preparations contracted by 60 mM KCl of five couples of benzanilide and thiobenzanilide derivatives with the same
aromatic substitution patterns and their potency index (PI = IC50 benzanilide/IC50 thiobenzanilide). The decrease in percent of the contraction force is semilogarithmically plotted on
the ordinate against the concentration in ␮M on the abscissa. Symbols represent the arithmetic means SEM from five experiments. Panel A shows compound 2 (᭹) and 13
(ꢁ), panel B 4 (ꢀ) and 16 (ꢁ), panel C 7 (ꢂ) and 22 (ꢃ), panel D 8 (ꢄ) and 15 (♦) and panel E 10 ( ) and 18 ( ).
both a Free-Wilson analysis and a combined Hansch/Free-Wilson
analysis with those compounds which fulfil the basic criteria for
Free-Wilson analysis. First, all compounds showing not the gen-
eral pattern of ring 1-C( X)-N-ring 2 were eliminated from the
data set (23–26, 31). Second, the data matrix was encoded assign-
ing each unique aromatic ring substitution pattern one indicator
variable, whereby ring 1 and ring 2 were treated differently. This
resulted in nine variables for ring 1 and ten variables for ring 2.
In addition, one column for X = S (yes = 1, else = 0) was introduced.
Third, to avoid single tones we eliminated all variables which were
present only once in the data set (e.g. 3,5-dimethoxy (20) or 3,4,5-
trimethoxy (21) at ring 1). This led to the removal of eight indicator
variables and subsequently five compounds (5, 6, 12, 20, 21) from
the data set. The final data set thus contained 21 compounds with
biological activities ranging from 0.1 (18) to 57.5 ␮M (29). Stepwise
backward variable selection revealed a final equation which con-
tains six indicator variables (Eq. (1)). The most prominent effect
seen is the negative contribution of 3-pyridyl as ring 1 and the
beneficial effect of the thioamide versus the amide. Remarkably,
also 4-methylthienyl,3,5-difluorophenyl and 2,4,6-trifluorophenyl
as ring 2 show positive contribution whereas 4-fluorophenyl as
ring 1 negatively contributes to activity. In order to relate this

G. Brunhofer et al. / European Journal of Pharmaceutical Sciences 42 (2011) 37–44
43
to potential physicochemical properties we included also molar
refractivity, log P and polar surface area into the calculations. Using
the physicochemical descriptors alone no statistically sound model
could beobtained(r2 = 0.34, q2 = 0.00). Combining both methodsled
to a slight improvement of the predictive power accompanied by a
moderate decrease of the r2 value. This indicates that the combined
model might be a bit more robust in predictions. However, it has
to be noted that the predictive power of both models as obtained
by leave one out cross validation is not at all satisfactory.
14
12
10
8
A
6
4
2
4. Conclusion
0
By optimizing the formerly as antispasmodic discovered ben-
zanilides we developed a series of thiobenzanilide derivatives with
improved spasmolytic activity (IC50) which have no significant
impact on inotropy and chronotropy up to a concentration of
100 ␮M. Structure–activity relationship study using Free-Wilson
analysis was performed. It clearly showed that a 3-pyridyl moi-
ety on the acidic side of the molecule had a negative contribution
to the biological activity. On the other hand, thioamide versus
amide influenced the desired spasmolytic activity in a positive
manner. The difference in the biological effect of amide and
thioamide is obviously demonstrated by pair wise comparison. The
thioamide compounds are 4–12 fold more active than their amide
counterparts. Actually, compound 18 showed a 325 times higher
spasmolytic activity than its amide derivative 10. We suppose that
the compounds mediated their biological activity via various mech-
anisms including activation of gralpha-adrenoceptors, inhibition
of histamine receptors and influence of Ca2+-channels. Activation
of KATP-channels and acetylcholine receptors does not participate
significantly. Due to these polypharmacological characteristics the
herein presented compounds could be applied in the treatment of
multifactorial diseases, e.g. the irritable bowl syndrome. Of course,
further biological investigation of this highly interesting compound
class is necessary.
14
12
10
8
B
histamine
compound 28 (2 µM)
**
6
**
4
**
2
*
0
Concentration histamine (µM)
10
phenylephrine
compound 28 (2 µM)
C
8
6
4
2
0
**
**
Acknowledgement
This study was supported in part by the Hochschuljubiläumss-
tiftung of the city Vienna.
References
Concentration phenylephrine (µM)
Apsel, B., Blair, J.A., Gonzalez, B., Nazif, T.M., Feldman, M.E., Aizenstein, B., Hoffman,
R., Williams, R.L., Shokat, K.M., Knight, Z.A., 2008. Targeted polypharmacology:
discovery of dual inhibitors of tyrosine and phosphoinositide kinases. Nat. Chem.
Biol. 4, 691–699.
Bolognesi, M.L., Cavalli, A., Melchiorre, C., 2009. Memoquin: a multi-target-directed
ligand as an innovative therapeutic opportunity for Alzheimer’s disease. Neu-
rotherapeutics 6, 152–162.
Brunhofer, G., Handler, N., Leisser, K., Studenik, C.R., Erker, T., 2008. Benzanilides
with spasmolytic activity: chemistry, pharmacology, and SAR. Bioorg. Med.
Chem. 16, 5974–5981.
Erker, T., Handler, N., Brunhofer, G., Rawnduzi, P., Studenik, C., 2007. Benzothioamide
und deren Verwendung. PCT/AT2007/000582.
16
14
12
10
8
D
acetylcholine
compound 28 (2 µM)
6
4
Estrada, S., Rojas, A., Mathison, Y., Israel, A., Mata, R., 1999. Nitric oxide/cGMP medi-
ates the spasmolytic action of 3,4ꢂ-dihydroxy-5,5ꢂ-dimethoxybibenzyl from
Scaphyglottis livida. Planta Med. 65, 109–114.
2
Galanski, M.E., Erker, T., Studenik, C.R., Kamyar, M., Rawnduzi, P., Pabstova, M.,
Lemmens-Gruber, R., 2005. Synthesis and pharmacological profile of non-
peptide vasopressin antagonists. Eur. J. Pharm. Sci. 24, 421–431.
Godfraind, T., Miller, R., Wibo, M., 1986. Calcium antagonism and calcium entry
blockade. Pharmacol. Rev. 38, 321–416.
Gutierrez, R., Solis, R., 2009. Relaxant and antispasmodic effects of extracts of the
orchid Encyclia michuacana on isolated guinea pig ileum. J. Nat. Med. 63, 65–68.
Handler, N., Brunhofer, G., Studenik, C., Leisser, K., Jaeger, W., Parth, S., Erker, T., 2007.
‘Bridged’ stilbene derivatives as selective cyclooxygenase-1 inhibitors. Bioorg.
Med. Chem. 15, 6109–6118.
0
Concentration acetylcholine µM
Fig. 2. Possible mode of action of compound 28 on terminal ileum preparations.
Panel A shows the effect of compound 28 on KATP channels in comparison with
glibenclamide (n = 3), panel B the effect on histamine-contracted preparations
(n = 9), panel C the effect on phenylephrine-contracted ilea (n = 3), and panel D the
effect on acetylcholine-contracted ilea (n = 3). Significance level: *P < 0.05; **P < 0.01.
Hernandez-Romero, Y., Rojas, J.-I., Castillo, R., Rojas, A., Mata, R., 2004. Spasmolytic
effects, mode of action, and structure–activity relationships of stilbenoids from
Nidema boothii. J. Nat. Prod. 67, 160–167.

44
G. Brunhofer et al. / European Journal of Pharmaceutical Sciences 42 (2011) 37–44
Hopkins, A.L., 2008. Network pharmacology: the next paradigm in drug discovery.
Nat. Chem. Biol. 4, 682–690.
Hopkins, A.L., Mason, J.S., Overington, J.P., 2006. Can we rationally design promis-
cuous drugs? Curr. Opin. Struct. Biol. 16, 127–136.
Rasko, D.A., Moreira, C.G., Li, D.R., Reading, N.C., Ritchie, J.M., Waldor, M.K., Williams,
N., Taussig, R., Wei, S., Roth, M., Hughes, D.T., Huntley, J.F., Fina, M.W., Falck,
J.R., Sperandio, V., 2008. Targeting QseC signaling and virulence for antibiotic
development. Science 321, 1078–1080.
Roth, B.L., Sheffler, D.J., Kroeze, W.K., 2004. Magic shotguns versus magic bullets:
selectively non-selective drugs for mood disorders and schizophrenia. Nat. Rev.
Drug Discov. 3, 353–359.
Shah, A.J., Gowani, A.S., Zuberi, A.J., Ghayur, M.N., Gilani, A.H., 2010. Antidiarrhoeal
and spasmolytic activities of the methanolic crude extract of Alstonia scholaris
L. are mediated through calcium channel blockade. Phytother. Res. 24, 28–32.