Simple N,N-dimethyl phenylsulfonamides show potent anticonvulsant effect in two standard epilepsy models

Article abstract of DOI:10.1016/j.bmcl.2016.11.035

Optimization of the previously reported benzothiazine analogue A led to the identification of compound 1, which showed anti-convulsant activity in two golden standard animal models of seizure, the MES and scPTZ models. Structure-activity relationship investigation of compound 1 revealed compounds 2, 6 and 19 as attractive anti-epileptic drug (AED) candidates with potent anticonvulsant effect in both the MES and scPTZ models. As these compounds are structurally different from existing AEDs, determination of their mechanism of actions could provide clues to understanding current therapy-resistant seizures. Moreover, these simple phenylsulfoneamide compounds could be good starting points for searching broad spectrum AEDs by such in vivo screening.

Full text of DOI:10.1016/j.bmcl.2016.11.035

Bioorganic & Medicinal Chemistry Letters 27 (2017) 94–97
Contents lists available at ScienceDirect
Bioorganic & Medicinal Chemistry Letters
journal homepage: www.elsevier.com/locate/bmcl
Simple N,N-dimethyl phenylsulfonamides show potent anticonvulsant
effect in two standard epilepsy models
⇑
Tomoyuki Tanaka , Nana Yajima, Tomoko Kiyoshi, Yoshiki Miura, Seiji Iwama
Drug Research Division, Sumitomo Dainippon Pharma Co., Ltd, 33-94 Enoki-cho, Suita, Osaka 564-0053, Japan
a r t i c l e i n f o
a b s t r a c t
Article history:
Received 29 July 2016
Revised 18 October 2016
Accepted 10 November 2016
Available online 15 November 2016
Optimization of the previously reported benzothiazine analogue A led to the identification of compound
1, which showed anti-convulsant activity in two golden standard animal models of seizure, the MES and
scPTZ models. Structure-activity relationship investigation of compound 1 revealed compounds 2, 6 and
19 as attractive anti-epileptic drug (AED) candidates with potent anticonvulsant effect in both the MES
and scPTZ models. As these compounds are structurally different from existing AEDs, determination of
their mechanism of actions could provide clues to understanding current therapy-resistant seizures.
Moreover, these simple phenylsulfoneamide compounds could be good starting points for searching
broad spectrum AEDs by such in vivo screening.
Keywords:
Phenylsulfoneamide
Epilepsy
In vivo screening
MES
Ó 2016 Elsevier Ltd. All rights reserved.
PTZ
Background
still provide a newly way to struggle current refractory epilepsy.8
(Fig. 1).
Epilepsy is a complex neurological disorder that affects about
50 million people worldwide.1 Epileptic seizures can be classified
as partial or generalized, and are difficult to control in up to 30%
of patients, despite optimal use of available anti-epileptic drugs
(AEDs).2,3 In addition to their limited efficacy, most AEDs pose
safety concerns, including CNS toxicity, drug-drug interaction, skin
rash, teratogenicity, and hepatotoxicity.4
Among seizure animal models, the maximal electroshock sei-
zure model (MES) and the subcutaneous pentylenetetrazol model
(scPTZ) are well-established, and have been used in a number of
studies associated with anticonvulsants drug discovery.5 Almost
all conventional AEDs (such as, Carbamazepine and Phenytoin)
and recently launched AEDs (such as, Topiramate and Lamotrigine)
produce potent anticonvulsant effect in MES model, but no or weak
efficacy in scPTZ model.4 These findings were clinically confirmed
as most AEDs show excellent efficacy for partial seizure, but have
only limited effect for generalized seizure. Thus, we hypothesized
that compounds that can act as potent anticonvulsants in both
MES and scPTZ models, would be useful for treatment of both par-
tial and generalized seizure.6 Recent AED discovery research
review article7 said this phenotypic screening nowadays seems
old-fashioned, identifying novel structural categories with broad
anti-epileptic spectrum by conventional in vivo screening might
We have previously found that the benzothiazine analogue A
has potent anticonvulsant effect in both MES and scPTZ models.10
However there is one negative point about this compound, its 4th
position may work as Michael acceptor. If this position is reacted
with proteins, the resultant modified protein sometime cause sev-
ere side effect such as liver dysfunction or skin lash. In order to
avoid this risk, we have searched new chemotypes. Further opti-
mization of compound A, in particular the thiazine ring, led to
the smaller compound 1 (MW 219.68) with potent anticonvulsant
effect in both MES model and scPTZ models. Having a smaller
molecular weight, compound 1 was considered ideal for good brain
penetration.
Although phenylsulfonamide scaffold is known as herbicides11
,
to the best of our knowledge, there are no reports that describe
them as anticonvulsants. Although several AEDs with a sulfon-
amide structure (e.g. Topiramate and Zonisamide) are clinically
available, their efficacy is limited to partial seizure, which corre-
lates with non-clinical data (Table 1). As compound 1 showed
weaker anticonvulsant activity in the scPTZ model than in
the MES model, our main objective was to improve its efficacy in
the scPTZ model, while keeping good anticonvulsant effect in the
MES model. Herein, we report the structure-activity relationship
of simple phenylsulfonamides and evaluate their anticonvulsant
activity in the MES and scPTZ models.
All compounds, except for 20, were prepared by reaction of the
corresponding sulfonyl chloride and amine in one step. Catalytic
⇑
Corresponding author.
E-mail address: tomoyuki-1-tanaka@ds-pharma.co.jp (T. Tanaka).
http://dx.doi.org/10.1016/j.bmcl.2016.11.035
0960-894X/Ó 2016 Elsevier Ltd. All rights reserved.

T. Tanaka et al. / Bioorganic & Medicinal Chemistry Letters 27 (2017) 94–97
95
in 2 by a pyrrolidine (compound 10) resulted in a complete loss of
anticonvulsant activity in the MES model. Introduction of a hydro-
xyl group into the terminal carbon of the alkyl chain (compound 7)
dramatically diminished the activity in both models. These find-
ings indicated that a sterically small alkyl-sulfonamide or an un-
substituted sulfonamide is the most suitable substituent.
Cl
S
O
Cl
S
N
H
CH3
CH3
N
N
O
CH3
CH3
O
O
O
MolecularWeight: 300.76
MolecularWeight: 219.68
1
A
O
O
O
SO2NH2
Next we focused on optimizing the substituent at the meta posi-
tion of the phenyl ring moiety. The results are summarized in
Table 3. The anticonvulsant effect of the unsubstituted compound
11 was weaker than that of compound 2 in both seizure models. As
for electron-withdrawing groups, the anticonvulsant effect of 14
with a fluorine atom, 15 with a bromine atom, 16 with a trifluo-
romethyl, 17 with a cyano, 18 with a nitro, and 19 with a trifluo-
romethoxy was almost comparable to that of 2 in the MES
model. Meanwhile, only 19 among these compounds showed an
anticonvulsant activity similar to that of 2 in the scPTZ model.
With regard to electron donating groups, obvious results were
obtained, i.e. 12 with a methyl group, 13 with a methoxy group,
20 with an amino group, and 21 with a phenoxy group all showed
no anticonvulsant effect in the MES model even at 100 mg/kg.
Taking the results of electron-withdrawing groups into considera-
tion, we concluded that the electron density of the phenyl moiety
greatly affects the anticonvulsant activity. Finally, we screened
three types of di-chlorinated phenyl sulfonamide derivatives for
the highest activity. Introduction of another chlorine atom on the
phenyl ring (compounds 22–24) resulted in diminished anticon-
vulsant activity in both the MES and scPTZ models. All compounds
did not show abnormal behaviors, such as somnolence, sedation or
muscle relaxation up to 100 mg/kg dosing. So we think these com-
pounds do not have neurotoxic potentials.
S
O
NH2
O
O
N
O
O
O
Topiramate
Zonisamide
Fig. 1. Chemical structure of the hit compound 1 as compared to that those of two
existing AEDs (Topiramate and Zonisamide).
Table 1
Anticonvulsant effect of 1 as compared to that of compound A and some existing
AEDs.
Dose (mg/kg)
MES
scPTZ
A
1
100
50
400
100
–
3/3a
3/3a
3/3a
2/3a
3/3a
3/3a
3/3a
1/3a
Topiramate
Zonisamide
33 mg/kgb
20 mg/kgc
>800 mg/kgb
>500 mg/kgc
–
a
b
c
Number of animals with no convulsion/test animals.
Data obtained from the literature (i.p. administration, Ref. 9).
ED50 values determined in our laboratory.
hydrogenation of compound 18 under H2 atmosphere afforded
compound 20 (Scheme 1).
When compound 2 was incubated with rat hepatic microsomes
at 37 °C for 30 min, 95% of its parent form remained in the solution,
indicating that this compound is metabolically stable and that its
anticonvulsant activity does not require the generation of com-
pound 6. These results indicate that the most suitable substituent
on the phenyl ring is a lipophilic electron-withdrawing group with
appropriate molecular size. A chlorine atom fulfilled these require-
ments. The ED50 of compound 2 on MES and scPTZ was calculated
approximately 12 and 34 mg/kg, respectively.
In conclusion, we found that simple, small-sized phenyl sulfon-
amides show anticonvulsant effect in two golden standards of sei-
zure animal models, i.e. the MES and scPTZ models. Optimization
of the alkyl chain on the nitrogen atom of these sulfonamides
revealed that dimethylsulfonamide is the most suitable amide
structure for anticonvulsant activity in both models. Introduction
First, we examined changes in the position of the chlorine atom
on the phenyl ring. The results are summarized in Table 2. Com-
pounds with the chlorine at the ortho (1), meta (2),12 or para (3)
position showed comparable anticonvulsant activity in the MES
model, although the meta (2) showed several fold stronger anticon-
vulsant activity than compounds 1 and 3 in the scPTZ model. Thus,
a chlorine at the meta position of the phenyl ring was considered as
ideal for further optimization. Insertion of a methylene unit
between the phenyl ring and the dimethyl sulfonamide (com-
pound 4) diminished the activity in the MES model. Compound
5, the monomethyl analogue of 2, showed slightly weaker activity
in the scPTZ, whereas the desmethyl analogue 6 showed an activity
similar to that of 2. The diethyl sulfonamide 8 and the methyl-
propyl sulfonamide 9 both showed weaker anticonvulsant effect
than 2 in the scPTZ model. Replacement of the dialkylamino group
1
2
3
: R1 = 2-Cl, : R1 = 3-Cl, : R1 = 4-Cl,
HNMe2
THF
R1
R1
11 : R1 = H, 12 : R1 = 3-Me, 13 : R1 = 3-OMe,
14 : R1 = 3-F, 15 : R1 = 3-Br, 16 : R1 = 3-CF ,
17 : R1 = 3-CN, 18 : R1 = 3-NO2, 19 : R1 = 33-OCF3
21 : R1 = 3-OPh
Cl
NMe2
S
S
O O
O O
Cl
Cl
4
6
8
10
,
5
,
: n = 1 R2 = R3 = Me; : n = 0 R2 = H, R3 = Me;
amine
O O
S
n
,
7 ,
9
O O
: n = 0 R2 = R3 = H;
: n = 0 R2 = Me, R3 = CH2CH2OH
: n = 0 R2 = R3 = Et; : n = 0 R2 = Me, R3
=
nPr
S
,
,
THF
NR2R3
Cl
n
,
: n = 0 NR2R3 = pyrrolidyl
NO2
NH2
H2, Pd / C
MeOH
NMe2
NMe2
S
S
O O
O O
18
20
Cl
Cl
22 : 3,4-dichloro
23 : 2,5-dichloro
HNMe2
THF
Cl
Cl
Cl
NMe2
24
: 3,5-dichloro
S
S
O O
O O
Scheme 1. Synthetic routes of the phenylsulfoneamide derivatives 1–24.

96
T. Tanaka et al. / Bioorganic & Medicinal Chemistry Letters 27 (2017) 94–97
Table 2
Table 3
Optimization of substituent position and sulfonamide moiety.
Effect of various substituents on anticonvulsant activity.
Compound
Structure
Cl
Dose (mg/kg)
MESa
scPTZa
Compound
Structure
Dose (mg/kg)
MESa
scPTZa
1
400
100
50
3/3
3/3
3/3
0/3
3/3
1/3
–
2
400
100
50
–
3/3
3/3
2/3
1/3
Cl
CH3
N
3/3
3/3
3/3
CH3
N
S
CH3
25
–
25
O
O
O
S
CH3
O
O
2
400
100
50
25
12.5
6.25
–
3/3
3/3
2/3
1/3
–
Cl
3/3
3/3
3/3
2/3
0/3
11
12
100
50
3/3
0/3
1/3
–
CH3
N
CH3
N
S
CH3
S
CH3
O
O
–
O
100
100
0/3
0/3
1/3
1/3
CH3
3
4
400
100
50
–
3/3
1/3
–
Cl
CH3
N
3/3
3/3
3/3
CH3
N
S
CH3
25
–
O
S
CH3
O
O
O
O
100
0/3
–
Cl
Cl
13
14
CH3
O
F
O
O
S
CH3
CH3
N
N
CH3
S
CH3
O
O
5
100
50
25
3/3
3/3
3/3
3/3
1/3
0/3
100
50
25
3/3
3/3
3/3
3/3
1/3
0/3
H
N
CH3
N
S
CH3
O
O
S
CH3
6
7
8
100
50
25
3/3
3/3
3/3
2/3
2/3
1/3
Cl
Cl
Cl
O
15
16
100
50
25
3/3
3/3
1/3
2/3
1/3
–
Br
F
NH2
CH3
N
S
S
O
O
S
CH3
O
F
100
0/3
1/3
O
100
50
25
3/3
2/3
0/3
3/3
0/3
–
CH3
N
F
OH
O
O
CH3
N
100
50
25
3/3
3/3
1/3
3/3
1/3
1/3
S
CH3
CH3
CH3
O
O
N
17
100
50
25
3/3
3/3
2/3
0/3
–
–
N
S
O
O
CH3
N
9
100
50
2/3
1/3
2/3
0/3
Cl
Cl
S
CH3
CH3
N
O
O
S
CH3
18
19
100
50
25
3/3
2/3
1/3
1/3
–
–
O
O
O
O
N
F
10
100
0/3
–
CH3
N
S
CH3
N
O
O
S
O
O
100
50
25
3/3
2/3
2/3
2/3
2/3
0/3
F
F
a
Efficacy was evaluated by the number of animals with no convulsion/test
animals.
O
CH3
N
S
CH3
of a lipophilic electron withdrawing group at the meta position of
the phenyl ring improved compounds anticonvulsant activity in
both seizure models. We believe that several compounds, such as
2, 6 and 19 are worthy of further evaluations in other seizure mod-
els, including the 6 Hz and rat amygdala kindling models.13 Deter-
mination of their mechanism of actions could provide clues to
understanding current therapy-resistant seizures.
O
O
20
100
0/3
–
NH2
CH3
N
S
CH3
O
O

T. Tanaka et al. / Bioorganic & Medicinal Chemistry Letters 27 (2017) 94–97
97
Table 3 (continued)
Acknowledgements
Compound
Structure
Dose (mg/kg)
100
MESa
0/3
scPTZa
–
We are grateful to Yoshie Nakatsuji for recording HRMS spectra,
and Keiko Bando for Elementary analysis. We also would like to
thank Yoshiaki Isobe for their support and encouragement.
21
O
References
CH3
N
1. FDA fact sheet: <http://www.who.int/mediacentre/factsheets/fs999/en/>.
2. Kwan P et al. Exp Rev Neurother. 2006;6:397.
S
CH3
O
O
3. Perucca E et al. Lancet Neurol. 2007;6:793.
4. Loscher W. Sezures. 2011;20:359.
5. Schmidt D et al. Epilepsy Res. 2002;50:71.
22
100
50
3/3
0/3
0/3
–
Cl
Cl
6. NIH once recommended these ‘‘in vivo screening” approach, see; Stables JP,
et al. ‘‘The NIH Anticonvulsant Drug Development (ADD) Program: Preclinical
Anticonvulsant Screening Project”. <http://www.ninds.nih.gov/research/asp/
addadd_review.pdf>.
7. Loscher W et al. Nat Rev Drug Discovery. 2013;12:757.
8. Recent in vivo screening research which succeeded in identifying new AED
scaffold, see Malik S. Eur J Med Chem. 2014;84:42.
CH3
N
S
CH3
O
O
23
24
100
0/3
–
Cl
Cl
9. Ucar H et al. J Med Chem. 1998;41:1138.
10. Tanaka T et al. Bioorg Med Chem Lett. 2015;25:4518.
11. US3518075 (Patent).
CH3
N
S
CH3
12. Spectral data for compound 2:1H NMR (CDCl3, 400 MHz) d 7.77–7.78 (1H, m),
7.66–7.68 (1H, m), 7.57–7.66 (1H, m), 7.48–7.52 (1H, m), 2.74 (6H, s); 13C NMR
(CDCl3, 400 MHz) d 137.4, 135.3, 132.8, 130.3, 127.7, 125.8, 37.9; HRMS (ESI+)
m/z calcd for C8H10NO2ClNaS (M+Na)+: 242.0013, found: 242.0017; Anal. Calcd
for C8H10ClNO2S; C, 43.74; H, 4.59; N, 6.38; S, 14.60; Cl, 16.14. Found: C, 43.85;
H, 4.57; N, 6.45; S, 14.59; Cl, 16.19.
O
O
100
50
25
3/3
2/3
0/3
1/3
–
–
Cl
CH3
N
13. Sato M et al. Electroenceph Clin Neurophysiol. 1990;76:459.
Cl
S
CH3
O
O
a
Efficacy was evaluated by the number of animals with no convulsion/test
animals.