Synthesis, anticonvulsant activity, and structure-activity relationships of sodium channel blocking 3-aminopyrroles

Article abstract of DOI:10.1021/jm970327j

Starting from the corresponding acetophenone and glycine derivatives, a series of new 3-aminopyrroles was synthesized in few steps. Using this procedure with hydrazine and hydroxylamine instead of the glycinates provides access to 3-aminopyrazoles and 5-a

Full text of DOI:10.1021/jm970327j

J . Med. Chem. 1998, 41, 63-73
63
Syn th esis, An ticon vu lsa n t Activity, a n d Str u ctu r e-Activity Rela tion sh ip s of
Sod iu m Ch a n n el Block in g 3-Am in op yr r oles
Klaus Unverferth,*,† J u¨rgen Engel,† Norbert Ho¨fgen,† Angelika Rostock,† Ralf Gu¨nther,† Hans-J oachim Lankau,†
Manfred Menzer,† Andreas Rolfs,‡ J u¨rgen Liebscher,‡ Birgit Mu¨ller,§ and Hans-J o¨rg Hofmann§
Corporate Research and Development ASTA Medica Group, Arzneimittelwerk Dresden GmbH, P.O. Box 010131,
D-01445 Radebeul, Germany, Humboldt University Berlin, Department of Chemistry, Hessische Strasse 1-2,
D-10115 Berlin, Germany, and University of Leipzig, Institute of Biochemistry, Faculty of Biological Sciences,
Pharmacy and Psychology, Talstrasse 33, D-04103 Leipzig, Germany
Received May 16, 1997X
Starting from the corresponding acetophenone and glycine derivatives, a series of new
3-aminopyrroles was synthesized in few steps. Using this procedure with hydrazine and
hydroxylamine instead of the glycinates provides access to 3-aminopyrazoles and 5-amino-1,2-
oxazoles. The various derivatives were tested for anticonvulsant activity in a variety of test
models. Several compounds exhibit considerable activity with a remarkable lack of neurotox-
icity. 4-(4-Bromophenyl)-3-morpholinopyrrole-2-carboxylic acid methyl ester, 3, proved to be
the most active compound. It was protective in the maximal electroshock seizure (MES) test
in rats with an oral ED50 of 2.5 mg/kg with no neurotoxicity noted at doses up to 500 mg/kg.
Compound 3 blocks sodium channels in a frequency-dependent manner. The essential
structural features which could be responsible for an interaction with an active site of the
voltage-dependent sodium channel are established within a suggested pharmacophore model.
Epilepsy affects 1% of the world’s population accord-
ing to epidemiological studies. Current clinically avail-
able drugs produce satisfactory seizure control in 60-
70% of patients.1 Nearly 95% of prescriptions written
by physicians worldwide for the treatment of epilepsy
are for the old antiepileptic drugs developed prior to
1975. During the past five years, several new drugs
have been approved or are in the process of being
approved, e.g., felbamate, lamotrigine, gabapentin, topi-
ramate, vigabatrin, and tiagabine.1 Although these
drugs have been shown to be effective in reducing
seizures in a number of patients, their efficacy does not
appear to be superior to that of the drugs developed
earlier. Thus, the triazine derivative lamotrigine shows
an efficacy rather similar to the well-known and world-
wide used dibenz[b,f]azepine derivative carbamazepine.2
Therefore, the need for more effective and less toxic
antiepileptic drugs still exists.3 Numerous derivatives
stemming from 5-, 6-, and 7-membered heterocycles
exert a more or less strong anticonvulsant activity.
Considering the various classes of heterocycles, it is
striking that only few representatives with remarkable
activity come from the pyrrole group.4 This report
describes the synthesis and anticonvulsant activity of
a series of 3-aminopyrroles. Some of these compounds
exhibit potent anticonvulsant activity and low minimal
neurotoxicity in several animal seizure models. They
are structurally different from the clinically effective
anticonvulsants as well as some recently published
aroyl(ω-aminoacyl)pyrroles.4
action was found for felbamate and topiramate. Topi-
ramate also affects chloride currents and increases the
number of channel openings induced by GABA. Con-
trary to this, lamotrigine acts by prolonging inactivation
of voltage-sensitive Na+ channels. Selectivity differ-
ences for the R subunits of Na+ channels might be
responsible for the two different mechanisms.5 The
3-aminopyrroles seem to act by inhibition of the sodium
channel in a frequency-dependent manner, too. On the
basis of this observation, the examination of structure-
activity relationships of 3-aminopyrroles and other well-
known voltage-dependent Na+ channel blockers might
serve to identify a common pharmacophore.
Ch em istr y
Synthetic routes to 3-aminopyrroles are scarce and
often limited to the synthesis of N-unsubstituted com-
pounds.6 Pyrroles with an N,N-disubstituted amino
group at 3-position have been synthesized by reaction
of 6-amino-1,3-thiazines with butyllithium,7 by interac-
tion of azirines with acetylene dicarboxylate,8 starting
from R-acylaminothioamides,9 by nucleophilic substitu-
tion of a nitro group in 3,4-dinitropyrroles by amines,10
by cyclization of a specific trimethinium salt,11 and
starting from 3-aminothioacrylamides and glycinates,12-14
respectively. Here, we present a synthetic route to
various new N,N-disubstituted 3-aminopyrroles II
(Scheme 1). This methodology is based on the last-
mentioned glycinate route and permits synthetic diver-
sity of products. The reaction of 3-aminothioacryla-
mides I with hydrazine and hydroxylamine instead of
the glycinates leads to 3-aminopyrazoles III and 5-amino-
1,2-oxazoles IV, respectively (Scheme 1).
Phenytoin and carbamazepine prolong the inactive
state of the voltage-dependent Na+ channels. A similar
* To whom correspondence should be addressed.
† Arzneimittelwerk Dresden.
‡ Humboldt University Berlin.
§ University of Leipzig.
X Abstract published in Advance ACS Abstracts, December 15, 1997.
S0022-2623(97)00327-0 CCC: $15.00 © 1998 American Chemical Society
Published on Web 01/01/1998

64 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 1
Unverferth et al.
Sch em e 1
with thionyl chloride. These intermediates are used to
obtain several amides (compounds 27-33 in Table 1).
2-Acetyl-4-(4-chlorophenyl)-3-morpholinopyrrole 34
can be synthesized by acylation of 25 with dimethylac-
etamide/POCl3 (Scheme 6). The 1-methyl-2-[2-meth-
oxycarbonylethyl]pyrrole 35 is obtained by alkylation
of 26 with methylacrylate in the presence of boron
trifluoride (Scheme 6).
Treatment of 2 with ethyl isocyanate furnished the
pyrroloimidazolinedione 36 (Scheme 7).
According to Scheme 1, 4-(4-chlorophenyl)-3-mor-
pholinopyrazole (38) can be obtained by ring closure of
compound 37 with hydrazine. Ring closure of 37 with
hydroxylamine provides 4-(4-chlorophenyl)-5-morpholino-
1,2-oxazole (39) (Scheme 8).
P h a r m a cology
The 3-aminopyrroles were tested for anticonvulsant
activity and minimal motor impairment. The tests used
were the maximal electroshock seizure (MES) test, the
pentylenetetrazole (PTZ) test, and the rotorod test,
respectively. These tests have been used to identify and
evaluate the anticonvulsant potential of screening sub-
stances.18 The obtained results are listed in Table 2.
In particular compound 3 potently protects mice in
the MES test after ip administration. The ED50 ) 27.8
mg/kg is comparable to the MES ED50 values of pheny-
toin and carbamazepine, respectively. Also in the MES
test in rats it is more effective than all other amino-
pyrrole derivatives and even more effective than pheny-
toin and carbamazepine after po administration. These
3-aminopyrroles were not effective against PTZ-induced
seizures. This profile indicates an effectiveness against
generalized seizures because the MES is a valuable tool
to identify drugs with efficacy against generalized
tonic-clonic and focal seizures.19
Sch em e 2
The 3-amino-2-arylthioacrylic acid amides I are easily
available by condensation of arylthioacetic amides with
triethoxymethane and amines15 or by other methods.16
The arylthioacetic amides can be obtained by Willgerodt-
Kindler reaction of acetophenones with sulfur and the
corresponding amine.17
The synthesis of 3-aminopyrroles is shown in Scheme
2. The first step is the substitution reaction of 3-amino-
2-arylthioacrylic acid amides with derivatives of glycine.
The following oxidation of the intermediates by I2 or Br2
(for technical use H2O2) provides 5-amino-4-arylisothia-
zolium salts, which can be isolated or directly trans-
formed into 3-aminopyrroles by elimination of sulfur.
A 1,3-thiazine could possibly be involved in the ring
transformation.13 Some of the physical properties of the
new 3-aminopyrroles are listed in Table 1 (compounds
1-12).
Reaction of 3-morpholino-4-arylpyrroles with alkyl
halides or acyl halides in the presence of sodium hydride
(Scheme 3) provides the 1-substituted compounds 13-
22 listed in Table 1.
The 4-(4-chlorophenyl)-3-morpholinopyrrole-2-carbox-
ylic acid methyl esters can be hydrolyzed by sodium
hydroxide in isopropyl alcohol to yield the free acids
after acidification. 4-(4-Chlorophenyl)-3-morpholino-
pyrroles (Scheme 4) are available by hydrolysis and
decarboxylation in water (compounds 23-26 in Table
1).
The motor function is only weakly impaired by 3 in
mice after ip administration. No impairment was
observed in rats (po) up to a dose of 500 mg/kg. The
spread between the median effective dose and the
median minimal motor impairing dose reflects the
favorable tolerability of 3 in comparison with phenytoin
and carbamazepine.
As a result of its promising pharmacological activity,
3 was evaluated in corneal kindling studies in rats. The
ability of a compound to reduce the seizure severity of
the kindled rat suggests that it may be especially useful
for the treatment of complex partial seizures.20,21 As
shown in Table 3, 3 was impressive in its ability to
reduce the seizure severity in the corneal kindled rat
model. After oral administration, it was as efficacious
as carbamazepine and more efficacious than phenytoin.
Compound 3 blocks sodium channels in a dose-
dependent manner.22 Due to its limited solubility, an
IC50 could not be calculated. To compare the frequency
dependence with phenytoin, concentrations of both
drugs were applied, yielding a 20% reduction of the
sodium current in the corresponding preparations.
These concentrations were 100 µM for phenytoin and
10 µM for 3. A more pronounced frequency potentiation
of the block could be seen in this case. A 10 Hz
stimulation provided a doubling of the blocking effect
by phenytoin and more than the 3-fold increase of the
blocking effect by 3 in comparison to a 1 Hz stimulation.
The 3-morpholino-4-arylpyrrole-2-carboxylic acids
(Scheme 5), e.g., 23, can be transformed into the
corresponding acid chloride hydrochlorides by treatment

Sodium Channel Blocking 3-Aminopyrroles
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 1 65
Ta ble 1. Physical Properties of 3-Aminopyrroles
compd
R1
R2
R3
R4
yield, %
mp,°C
log Pa
formula
1b
2
3
H
H
H
H
H
H
H
H
H
H
H
H
COOCH3
N(CH2CH2)2O
N(CH2CH2)2O
N(CH2CH2)2O
N(CH2CH2)2O
N(CH2CH2)2O
N(CH2CH2)2O
N(CH2CH2)2O
N(C2H5)2
4-Cl
4-Cl
4-Br
4-Cl
3-Br
2-CH3
4-Cl
4-Cl
4-Cl
4-Cl
4-Cl
87
55
72
62
67
91
80
64
74
45
80
91
89
90
67
50
62
68
78
79
76
59
87
59
95
73
75
54
61
49
63
70
74
51
38
185-187
195-197
176-177
221-222
168-169
138
224-225
92-93
122-124
173-175
239-241
138
3.5
3.9
3.8
3.5
C16H17ClN2O3
C17H19ClN2O3
C16H17BrN2O3
C15H14ClN3O
C16H17BrN2O3
C17H20N2O3
C19H23ClN2O3
C17H21ClN2O2
C17H19ClN2O2
C22H22ClN3O2
C18H22ClN3O2
C17H22ClN3O2
C17H20N2O3
COOC2H5
COOCH3
CN
4
5
COOCH3
COOCH3
COOCH2CH(CH3)2
COOC2H5
COOCH3
COOCH3
COOC2H5
COOCH3
COOCH3
COOCH3
COOCH3
COOCH3
COOCH3
COOCH3
COOCH3
COOCH3
6c
2.8
7
8
9
N(CH2)5
N(CH2CH2)2NC6H5
N(CH2CH2)2NCH3
10
11b
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
4.4
3.3
N(CH3)CH2CH2N(CH3)2 4-Cl
CH3
CH3
CH2C6H5
COCH3
N(CH2CH2)2O
N(CH2CH2)2O
N(CH2CH2)2O
N(CH2CH2)2O
N(CH2CH2)2O
N(CH2CH2)2O
N(CH2CH2)2O
N(CH2CH2)2O
N(CH2CH2)2O
N(CH2CH2)2O
N(CH2CH2)2O
N(CH2CH2)2O
N(CH2CH2)2O
N(CH2CH2)2O
N(CH2CH2)2O
N(CH2CH2)2O
N(CH2CH2)2O
H
4-Cl
H
86-88
113-115
113-114
132
C17H19ClN2O3
C23H24N2O3
C18H19FN2O4
C18H19ClN2O4
C22H28N2O4
4-F
COCH3
4-Cl
4-C2H5
4-Cl
4-Cl
4-Cl
4-Cl
4-Cl
4-Cl
4-Cl
4-Cl
4-Cl
4-Cl
4-Cl
4-Cl
4-Cl
4-Cl
4-Cl
4-Cl
4-Cl
128-130
118-119
134
3.5
5.5
COC3H7
COC6H5
COOC6H5
CON(CH2CH2)2O COOCH3
SO2C2H5
H
CH3
H
CH3
H
H
H
H
H
H
H
H
CH3
C23H21ClN2O4
C23H21ClN2O5
C21H24ClN3O5
C18H21ClN2O5S
C15H15ClN2O3
C16H17ClN2O3
C14H15ClN2O
C15H17ClN2O
C15H16ClN3O2
C18H22ClN3O2
C18H20ClN3O2
C18H22ClN3O3
C17H20ClN3O2
C19H22ClN3O3
C20H25ClN4O2
C16H17ClN2O2
C19H23ClN2O3
146
194-196
166-168
147-148 dec
116-119 dec
130-131
92-93
275-277
230-233
237-239
204-205
248-250
249-250
254-257
208-209
116-118
3.1
4.2
1.8
2.3
3.2
4.2
2.4
4.2
3.7
3.3
3.1
3.1
3.3
COOCH3
COOH
COOH
H
H
CONH2
CONHC3H7
CONHCH2CHdCH2
CONHCH2CH2OCH3 N(CH2CH2)2O
CON(CH3)2
CON(CH2CH2)2O
CON(CH2CH2)2NCH3 N(CH2CH2)2O
COCH3
N(CH2CH2)2O
N(CH2CH2)2O
N(CH2CH2)2O
N(CH2CH2)2O
CH2CH2COOCH3
a
b
Experimental partition coefficient 1-octanol/water; log P (carbamazepine) ) 1.6 (this work); 1.76 (ref 56). See ref 12. c See ref 13.
Sch em e 3
Sch em e 4
Str u ctu r e-Activity Rela tion sh ip s
The activity data of the aminopyrrole series (Table
2) is derived from an in vivo assay and may, therefore,
be influenced by compound differences in bioavailability,
metabolism, blood-brain barrier penetration, etc. Nev-
ertheless, a tentative comparison of the compound
structures and activities may be helpful to get an idea
of structure-activity relationships. Thus, structures
with an ester moiety (2, 3, and 5) or keto group (34)
are very active, whereas the nitrile group, 4, reduces
the activity. The large size of the alkyl side chain of
the ester group in 7 or the alkyl spacer between the
ester group and the pyrrole ring in 35 may be respon-
sible for the loss of activity. However, absence of the
ester group as in the unsubstituted 25 did not result in
the loss of activity as also indicated by the activity of
the isosteric pyrazole derivative, 38, whereas loss of the
hydrogen donor function in the 1,2-oxazole derivative
39 and in the imidazolone derivative 40 causes loss of
activity (Figure 1). Substituents larger than the mor-
pholino group decrease the activity as indicated for 10,

66 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 1
Unverferth et al.
Ta ble 2. Anticonvulsant Activity and Minimal Motor
Sch em e 5
Impairment of 3-Aminopyrrole Derivativesa
MES
Rotorod
mice ip
56.4
compd
mice ip
58.1
(45.3-69.2) 6.8-18.2)
49.9 7.9
(36.0-68.8) (5.4-11.2)
27.8 2.5
(27.2-28.6) (1.2-4.2)
rat po
rat po
>500
1
2
3
11.4
(31.2-89.2)
158
(106-215)
166
>500
>500
(105-215)
>300
>300
>100
NE
4
5
30% 100
100% 100
60% 100
NE
100% 50
75% 50
50% 50
NE
>50
>50
>50
NE
Sch em e 6
6
7
10
175
12.5
>500
>500
(145-209)
100% 30
100% 30
33.1
(6.7-20.5)
100% 30
100% 50
13.6
11
12
17
<25
<30
35.4
<25
>50
>250
(24.1-37.5) (10.9-16.8) (22.6-53.6)
43.7 14.0 175
(35.9-52.7) (11.0-18.8) (133-215)
19
>500
22
23
25
NE
NE
NE
NE
>25
>200
NE
62.1
75% 25
17.7
>300
115
(53.8-69.8) (10.6-26.8) (86.9-150)
27
65.0
50% 50
>500
>50
(57.9-73.5)
NE
72.6
33
34
50% 50
10.6
>300
335
>50
>100
(59.9-92.0) (6.7-16.1)
(268-486)
NE
NE
35
36
38
NE
NE
NE
NE
>50
Sch em e 7
NE
NE
47.0
100% 50
100% 300
(40.0-57.0)
NE
39
NE
NE
NE
phenytoin
6.5
23.2
42.8
>500
(5.7-7.2)
carbamazepine 9.9
(8.8-10.7)
(21.4-25.4) (36.4-47.5)
11.9
47.8
361
(8.5-15.0)
(39.2-59.2) (319-402)
a
ED50 and rotorod in mg/kg; protection percentage at the
indicated dose in mg/kg, NE ) no effective response.
Ta ble 3. Anticonvulsant Activity in Corneally Kindled Rats
compound
route
po
dose, mg/kg
seizure score
Sch em e 8
3
25
50
75
100
10
25
50
75
100
150
10
4.3 ( 0.6
3.7 ( 0.6
1.7 ( 0.5a
1.9 ( 0.9a
5.0 ( 0.0
4.7 ( 0.2
4.4 ( 0.2
3.7 ( 0.7
3.0 ( 1.0a
3.0 ( 1.0a
4.8 ( 0.2
4.5 ( 0.2
3.8 ( 0.6
2.0 ( 0.8
phenytoin
ip
carbamazepine
po
20
40
80
a
p < 0.05.
It may be tempting to compare the structures of the
aminopyrroles and other molecules with anticonvulsant
activity to find out the structural elements essential for
action. In the past, several attempts were made to
postulate a general pharmacophore for the different
anticonvulsant classes, e.g., benzodiazepines,24 barbi-
turates,25 triazolines,26 and enaminones, respectively,27-29
and also for structurally different compounds with
similar anticonvulsant profiles.30-34 The various pos-
tulated pharmacophore models show no uniform picture.
Nevertheless, the presence of at least one aryl unit, one
or two electron donor atoms, and/or an NH group in a
while the replacement of the morpholino group by
N-methylpiperazine as in 11 increases the neurotoxicity.
In the case of the derivatives 17 and 19, it is
postulated that the active compound 1 (Table 2) is
responsible for the activity which might be formed by
metabolism, which seems to be excluded for 22. It is
known that N-acylpyrroles are rather unstable. An
X-ray study23 on 23 indicates a hydrogen bond between
the carboxy group and the morpholine ring which might
influence the structure and activity of this compound.

Sodium Channel Blocking 3-Aminopyrroles
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 1 67
F igu r e 1. Structures of selected active and nonactive derivatives (for discussion, see text).
F igu r e 2. Selected anticonvulsants for the development of a pharmacophore model. The essential structure elements are indicated
by dotted rectangles (R, hydrophobic unit; D, electron donor group; HAD, hydrogen donor/acceptor unit).
special spatial arrangement seems to be recom-
mended.30-34 On the basis of some ideas of Camerman
and Camerman30 and Wong et al.,31 respectively, J ones
and Woodbury32,33 defined a model with two electron
donors in some proximity to a bulky hydrophobic moiety.
Selecting other compounds as those of J ones and Wood-
bury, Codding et al.34 postulated a pharmacophore
consisting of a linear arrangement of a rotated phenyl
ring, an electron donor atom, and a hydrogen donor site,
which partially agrees with the model of J ones and
Woodbury. Brouillette et al.35-37 investigated the so-
dium channel blocking activity of several mono- and
bicyclic phenytoin analogues and concluded that high
log P values, a free imide group, and a specific aromatic
ring orientation are optimal for high binding affinity to
the sodium channel. These criteria were well fulfilled
in monocyclic hydantoins. However, the conclusions of
these studies were not related to other substance classes
acting at the same receptor site. Since only the primary
structures of several types of voltage-dependent sodium
channels are known,38-42 a study of the various anti-
convulsants with sodium channel blockade activity may
define structural elements which are essential for
activity. For this purpose, the five well-known and
structurally different compounds with anticonvulsant
activity, carbamazepine (CBZ), phenytoin (DPH), lam-
otrigine (LAM), zonisamide (ZON), and rufinamide
(CGP), were selected (Figure 2). These compounds show
a behavior similar to the most active aminopyrrole 3 in
animal screening tests and, moreover, act according to
the same mechanism.43-47
The five molecules have at least one aryl ring (R), one
electron donor atom (D), and a second donor atom in
close proximity to the NH group forming a hydrogen
bond acceptor/donor unit (HAD). In most cases this is
an amide bond. In Figure 2 these moieties are marked
by dotted rectangles. Most compounds are able to
realize two alternative conformational orientations of
the HAD unit, the one with the H acceptor function, the
other with the H donor function toward the aryl ring.
Only with phenytoin is it impossible to orient the H
acceptor function in this direction, and in lamotrigine
the H donor function cannot point to the aryl ring
because of missing flexibility. If this HAD unit should
be in fact essential for sodium channel-blocking activity,
the receptor site for this group seems to be rather
flexible.
In an initial study, a superimposition of all five
molecules was generated by a torsional flexible fit
procedure with a maximum overlap for the aryl rings
and the HAD parts, where the donor and acceptor parts
of the latter group were exchangeable. It can be seen
that also one electron donor atom of each compound can
be positioned in about the same region with deviations
smaller than 0.7 Å (Figure 3). All molecules are
principally able to realize conformers which bring the
above-mentioned groups closely together.
Of course, the various molecules may not be in
energetically favorable arrangements in this superim-
position. Therefore, calculations on the basis of ab initio
and semiempirical MO theory, molecular mechanics,
and dynamics were performed to obtain an overview on
the conformational properties of the various molecules,
their minimum conformations, and their flexibilities. On
the basis of these data, the distance relations between
the selected structural elements may be analyzed to
define the pharmacophore model more precisely.
Table 4 shows the average distances between the
various groups postulated as essential for action at
various levels of theory. It is obvious that the distance
ranges between the electron donor atom D and the aryl
ring R on one hand and between D and the HAD unit
on the other hand are very small at all levels of theory
in comparison to a larger distance range between R and
HAD. The average values and standard deviations
differ insignificantly depending on the different calcula-
tion methods. In comparison with the ab initio MO
data, the molecular mechanics results obtained with the
CHARMm force field reflect rather satisfactorily the
relationships between the essential structural elements.
It should be mentioned that the distance ranges ob-
tained by molecular dynamics are also relatively small.
On the basis of the molecular dynamics distance esti-
mations, the suggested pharmacophore model for com-
pounds acting as blockers of the voltage-dependent
sodium channel is visualized in Figure 4. This model
comprises an electron donor D in relatively limited
distance ranges of 3.2-5.1 Å to an aryl ring or other

68 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 1
Unverferth et al.
F igu r e 3. Two sights of a superimposition of the anticonvulsants carbamazepine (CBZ), phenytoin (DPH), lamotrigine (LAM),
zonisamide (ZON), and rufinamide (CGP) as result of a torsional flexible fit to reach maximum overlap between the R and HAD
units.
Ta ble 4. Distance Ranges between the Essential Structure
Elements R, D, and HAD of Five Anticonvulsantsa at Various
Levels of Theory
Ta ble 5. Calculated Distance Ranges between the
Pharmacophore Elements R, D, and HAD for Four
Anticonvulsantsa from Molecular Dynamics Simulations
methodb
R-HADc
R-Dc
D-HADc
compound
R-HADb
R-Db
D-HADb
MM
MD
AM1
PM3
5.85 ( 1.20
5.79 ( 1.17
5.78 ( 1.04
6.06 ( 1.04
5.67 ( 1.02
3.82 ( 0.26
3.88 ( 0.29
3.91 ( 0.38
3.93 ( 0.39
3.95 ( 0.43
4.56 ( 0.15
4.57 ( 0.19
4.75 ( 0.59
4.78 ( 0.32
4.64 ( 0.20
3
6.59 ( 0.06
4.88 ( 0.15
6.78 ( 0.69
4.47 ( 0.23
4.21 ( 0.11
5.45 ( 0.07
3.96 ( 0.27
4.69 ( 0.61
4.23 ( 0.06
5.28 ( 0.08
3.48 ( 0.07
4.32 ( 0.14
vinpocetine
dezinamide
remacemide
HF/3-21G
a
b
See Figure 5. In angstroms.
a
b
See Figure 2. MM: molecular mechanics, CHARMm force
field. MD: molecular dynamics, CHARMm force field. AM1. PM3:
semiempirical methods. HF/3-21G: ab initio MO theory. c Dis-
tances in angstroms.
F igu r e 4. Suggested pharmacophore model for anticonvul-
sants acting at the voltage-dependent sodium channel on the
basis of molecular dynamics simulations on carbamazepine,
phenytoin, lamotrigine, zonisamide, and rufinamide.
hydrophobic units R and of 3.9-5.5 Å to a hydrogen
bond acceptor/donor unit. The distance between R and
HAD spans a wider range of 4.2-8.5 Å. The hydropho-
bic unit R is not oriented in the same plane like the
other essential elements. In contrast to Brouillette et
al.,35-37 we could not find an indication for a specific
orientation of the aromatic ring. The rings are rotated
in relation to the R-D-HAD plane by 10-40°. In some
cases even free rotation of the phenyl ring is possible.
Now, it may be interesting to examine whether the
aminopyrroles reflect the conditions of the derived
pharmacophore model. Our analyses of the distance
relationships show that compound 3 fulfills the essential
demands of our pharmacophore. The agreement of the
calculated distances with the postulated pharmacophore
values is excellent (Table 5). To support the suggested
pharmacophore model, we considered some more struc-
turally different substances with anticonvulsant activity
for which the action as potent blockers of the voltage-
dependent sodium channel was recently recognized as
F igu r e 5. Structure of four anticonvulsants from different
structure classes fulfilling the demands of the general phar-
macophore model of Figure 4.
rather probable (Figure 5).47,48 All of these structurally
very dissimilar compounds contain aryl rings and
electron donor and H-bond donor/acceptor functions.
Vinpocetine and various conformations of remacemide
hydrochloride and dezinamide fulfill the essential de-
mands of our pharmacophore (Table 5). In the case of
vinpocetine only the distance R-D is too long, but it is
difficult to specify the aryl unit in this molecule. In
agreement with all other cases, we determined the
center of the phenyl rings as reference point for R, while
in vinpocetine the entire hydrophobic unit is more
extended. If we would select, for instance, the indole
portion as reference point, the distance between the aryl
unit and the electron donor would be shorter and the

Sodium Channel Blocking 3-Aminopyrroles
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 1 69
2-(4-Ch lor op h en yl)-3-[[(Eth oxyca r bon yl)m eth yl]a m i-
n o]th ioa cr ylic a cid m or p h olid e: mp 145 °C; IR (KBr) 3399,
1733; 13C NMR (DMSO-d6) δ 136.93 (CH), 170.38 (CdO),
10296.96 (CdS). Anal. (C17H20ClN2O3S) C, H, N.
2-(4-Br om op h en yl)-3-[[(m eth oxyca r bon yl)m eth yl]a m i-
n o]th ioa cr ylic a cid m or p h olid e: mp 161 °C; IR (KBr) 3382,
1746; 13C NMR (DMSO-d6) δ 136.71 (CH), 170.79 (CdO),
196.74 (CdS). Anal. (C16H18BrN2O3S) C, H, N.
Gen er a l Rin g Closu r e P r oced u r e. Meth od A. A solu-
tion of 0.08 mol of the 3-(alkoxycarbonyl)-2-arylthioacrylic acid
amides and 20.5 g (0.08 mol) iodine in 630 mL of ethanol was
stirred and heated to 40 °C. Fourteen milliliters (0.1 mol) of
triethylamine was added dropwise to this reaction mixture.
After 2 h of stirring at 40 °C, another 14 mL (0.1 mol) of
triethylamine was added. Two hundred and fifty milliliters
of the solvent was distilled off and the hot solution was filtered.
The clear solution was cooled in ice-water for 4 h. The solid
was collected by filtration and purified by recrystallization.
Method A was used for the preparation of the following
compounds.
4-(Ch lor op h en yl)-3-m or p h olin op yr r ole-2-ca r b oxylic
a cid m eth yl ester (1): IR (KBr) 3230, 1720; 1H NMR (CDCl3)
3.07 (t, NCH2CH2O), 3.72 (t, NCH2CH2O), 3.86 (s, OCH3), 6.85
(d, J ) 3.42, C5-H), 7.26-7.59 (m, Ph-H), 8.87 (s, N-H); 13C
NMR (CDCl3) 51.12 (OCH3), 51.48 (NCH2CH2O), 67.52
(NCH2CH2O), 114.88 (C4), 120.48 (C3), 122.17 (C5), 126.31
(CP h ), 127.96 (CP h ), 128.49 (CP h ), 134.79 (CP h ), 140.91 (C2),
160.51 (CdO). Anal. C, H, Cl, N.
requirements of the pharmacophore could be fulfilled.
Remacemide contains the HAD function as an amide
bond, but not at the correct position. If we take the
carbonyl oxygen as donor atom D, the hydrogen-bond
function will only be represented by the amine function.
In dezinamide the HAD unit is correctly placed, but the
calculated distance D-HAD is not in the postulated
range. The choice of the other oxygen atom as electron
donor leads to an increase in the corresponding distance
and a shortening of the distance R-D. This result
indicates that the distance range for D-HAD could
possibly be larger than originally suggested. Possibly,
the presence of only one component of the HAD unit at
the postulated position, e.g., only the H donor part, as
it is also postulated by other authors,30-37 in addition
to the two other essential structure elements R and D
may be sufficient for activity as blocker of the voltage-
dependent sodium channel. This is supported by the
fact, that compound 25 which has only the H donor part
of the HAD unit shows significant anticonvulsant activ-
ity, whereas 39 and 40 having only H acceptor groups
are inactive (Figure 1).
Exp er im en ta l Section
Ch em istr y. All melting points were determined on a
Boetius melting point apparatus PHMK 05. They are uncor-
rected. The IR spectra were registered on a Perkin-Elmer
1725x spectrometer. All absorption values are expressed in
wavenumbers (cm-1). Proton (1H NMR) and carbon (13C NMR)
nuclear magnetic resonance spectra were recorded on a Bruker
ARX 300 NMR spectrometer. Chemical shifts (δ) are in parts
per million (ppm) relative to Si(CH3)4 and coupling constants
(J ) are in hertz. The partition coefficients given as log P values
for 1-octanol/water were determined according to Yamagami
and Takao.49
2-Cya n o-4-(4-ch lor op h en yl)-3-m or p h olin op yr r ole (4).
1
IR (KBr) 3340, 2200; H NMR ((CD3)2CO) 7.24 (d, J ) 3.42,
C5-H), 10.89 (s, N-H); 13C NMR (DMSO-d6) 115.89 (C-Cl),
132.96 (CN), 143.67 (C2). Anal. C, H, Cl, N.
4-(3-Br om op h en yl)-3-m or p h olin op yr r ole-2-ca r b oxy-
lic a cid m eth yl ester (5): IR (KBr) 3430, 1695; 1H NMR
(CDCl3) 3.87 (s, OCH3), 9.06 (s, N-H); 13C NMR (CDCl3) 136.80
(C-Br), 160.40 (CdO). Anal. C, H, Br, N.
4-(2-Met h ylp h en yl)-3-m or p h olin op yr r ole-2-ca r b oxy-
lic a cid m eth yl ester (6): IR (KBr) 3260, 1710; 1H NMR
(CDCl3) 2.24 (s, CH3), 6.64 (d, J ) 3.35, C5H); 13C NMR (CDCl3)
20.65 (CH3). Anal. C, H, N.
Key-In t er m ed ia t es: 3-Am in o-2-a r ylt h ioa cr ylic Acid
Am id es. The synthesis of the 3-amino-2-aryl-thioacrylic acid
amides is described in ref 15.
4-(4-Ch lor op h en yl)-3-m or p h olin op yr r ole-2-ca r b oxy-
1
lic a cid isobu tyl ester (7): IR (KBr) 3300, 1700; H NMR
2-(4-Ch lor oph en yl)-3-m or ph olin oth ioacr ylic Acid Mor -
p h olid e (37). A solution of 290 g (1.13 mol) of (4-chlorophe-
nyl)thioacetic acid morpholide, 300 g (3.34 mol) morpholine
and 942 mL (5.67 mol) triethoxymethane was stirred and
heated at a bath temperature of 180-190 °C. [(4-chlorophe-
nyl)thioacetic acid morpholide:17 260 g (1.68 mol) of 4-chloro-
acetophenone, 293 g (3.36 mol) of morpholine, 140 mL of
methanol, and 1.6 g of p-toluenesulfonic acid hydrate were
stirred and heated under reflux for 9 h. After the mixture
was cooled to 65 °C, 5.0 L of methanol was quickly added
dropwise to the reaction mixture. At room temperature a solid
precipitated which was collected by filtration, washed four
times with methanol, and recrystallized from methanol, yield-
ing 360 g (83%), mp 96-99 °C.] Over a period of 8 h, 385 mL
of ethanol was distilled off. The bath temperature was then
decreased to 120 °C, and the remaining triethoxymethane was
removed by distillation. After the residue was cooled to 80
°C, 110 mL of ethanol was quickly added dropwise to the
mixture, which was further cooled to room temperature. The
precipitate was washed three times with 100 mL of ethanol
and recrystallized from butanol: yield 300 g (75%); mp 178-
181 °C; IR (KBr) 2966, 2849, 1615, 1114; 13C NMR (DMSO-
d6) δ 137.32 (CH), 198.45 (CdS). Anal. (C17H21ClN2O2S) C,
H, Cl, N.
(CDCl3) 1.02 (d, J ) 6.5 Hz, CH3), 2.09 (m, CH), 4.09 (d, J )
6.7, OCH2), 6.87 (d, J ) 3.5, C5-H); 13C NMR (CDCl3) 19.29
(CH3), 27.91 (CH), 59.94 (OCH2). Anal. C, H, Cl, N.
4-(4-Ch lor op h en yl)-3-(d ieth yla m in o)p yr r ole-2-ca r box-
1
ylic a cid eth yl ester (8): H NMR (CDCl3) 1.00 (t, J ) 7.1,
CH3), 1.41 (t, J ) 7.1, CH3), 3.15 (q, J ) 7.1, NCH2), 4.37 (q,
J ) 7.1, OCH2); 13C NMR (CDCl3) 13.63 (CH3), 14.56 (CH3),
47.57 (NCH2), 60.18 (OCH2) 133.66 (C-Cl). Anal. C, H, Cl,
N.
4-(4-Ch lor op h en yl)-3-p ip er id in op yr r ole-2-ca r b oxylic
a cid m eth yl ester (9): IR (KBr) 3280, 1665; 1H NMR (CDCl3)
1.53 (m, CH2), 2.99 (m, NCH2); 13C NMR (CDCl3) 24.15 (CH2),
26.57 (CH2), 52.40 (NCH2). Anal. C, H, Cl, N.
4-(4-Ch lor op h en yl)-3-(4-p h en ylp ip er a zin o)p yr r ole-2-
ca r boxylic a cid m eth yl ester (10): IR (KBr) 3395, 1704;
1H NMR (DMSO-d6) 3.29 (s, NCH2), 3.76 (s, OCH3); 13C NMR
(DMSO-d6) 49.06 (N-CH2), 50.11 (CH2N). Anal. C, H, Cl, N.
4-(4-Ch lor op h en yl)-3-(4-m et h ylp ip er a zin o)p yr r ole-2-
1
ca r boxylic a cid eth yl ester (11): IR (KBr) 3250, 1700; H
NMR (CDCl3) 1.37 (t, J ) 7.1, CH3), 2.29 (s, NCH3), 2.37 (m,
NCH2), 3.17 (m, NCH2), 4.35 (q, J ) 7.1, OCH2); 13C NMR
(CDCl3) 14.68 (CH3), 46.45 (NCH3), 51.34 (NCH2), 55.83
(NCH2), 60.13 (OCH2). Anal. C, H, Cl, N.
4-(4-Ch lor op h en yl)-3-[N-m eth yl-N-[2-(d im eth yla m in o)-
et h yl]a m in o]p yr r ole-2-ca r b oxylic a cid m et h yl est er
(12): IR (KBr) 3420, 1710; 1H NMR (CDCl3) 2.14 (s, N(CH3)2),
2.24 (m, CH2N), 2.79 (s, NCH3), 3.16 (m, NCH2); 13C NMR
(CDCl3) 41.94 (NCH3), 45.65 (N(CH3)2), 54.38 (NCH2), 57.85
(NCH2). Anal. C, H, Cl, N.
Syn th esis of 3-Am in op yr r oles 1-12. 3-[[(Alk oxyca r -
bon yl)m eth yl]a m in o]-2-a r ylth ioa cr ylic Acid Am id es. A
mixture of 0.2 mol of 3-amino-2-arylthioacrylic acid amides,
0.22 mol of aminoacetic acid esters, and 8 mL of triethylamine
in 600 mL of toluene was stirred and heated under reflux for
1 h. Triethylammonium chloride was filtered off, and the
solution was cooled overnight. The precipitated solid was
collected by filtration and purified by recrystallization from
methanol.
Meth od B. A solution of 0.35 mol of the 3-[[(alkoxycarbo-
nyl)methyl]amino]-2-arylthioacrylic acid amides in 1.33 L of
methanol was stirred and heated to reflux. Fifty-two and one-

70 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 1
Unverferth et al.
half milliliters (0.52 mol) of hydrogen peroxide (30%) was
slowly added to this solution within 30 min. An orange-colored
solution resulted, and the product precipitated within a few
minutes. The suspension was refluxed for 15 min and then
cooled in ice-water for 2 h. The solid was collected by
filtration and purified by recrystallization.
4-(4-Ch lor op h en yl)-3-m or p h olin o-1-(p h en yloxyca r bo-
n yl)p yr r ole-2-ca r boxylic a cid m eth yl ester (20): IR (KBr)
1765, 1730; 1H NMR (CDCl3) 3.86 (s, OCH3); 13C NMR (CDCl3)
52.74 (CH3), 150.19 (CdO), 163.53 (CdO). Anal. C, H, Cl,
N.
4-(4-Ch lor op h en yl)-3-m or p h olin o-1-(m or p h olin oca r -
bon yl)p yr r ole-2-ca r boxylic a cid m eth yl ester (21): 1H
NMR (DMSO-d6) 3.00 (m, NCH2), 3.45 (m, NCH2), 3.65 (m,
CH2O), 3.82 (s, CH3); 13C NMR (DMSO-d6) 45.45 (NCH2), 51.03
(NCH2), 65.50 (CH2O), 66.56 (CH2O), 131.23 (C-Cl), 151.53
(CdO), 159.61 (CdO). Anal. C, H, Cl, N.
4-(4-Ch lor op h en yl)-1-(eth ylsu lfon yl)-3-m or p h olin op y-
r r ole-2-car boxylic acid m eth yl ester (22): 1H NMR (CDCl3)
1.40 (t, CH3), 3.80 (q, SCH2); 13C NMR (CDCl3) 8.04 (CH3),
50.02 (SCH2), 116.4087 (C2), 123.85 (C5). Anal. C, H, Cl, N,
S.
Method B was used for the preparation of the following
compounds.
4-(4-Ch lor op h en yl)-3-m or p h olin op yr r ole-2-ca r b oxy-
lic a cid eth yl ester (2): IR (KBr) 3300, 1698; 1H NMR
(CDCl3) 1.38 (t, J ) 7.1, CH3), 4.34 (q, J ) 7.1, OCH2); 13C
NMR (DMSO-d6) 14.32 (CH3), 59.41 (OCH2), 130.16 (C-Cl).
Anal. C, H, Cl, N.
4-(4-Br om op h en yl)-3-m or p h olin op yr r ole-2-ca r b oxy-
lic a cid m eth yl ester (3): IR (KBr) 3300, 1719; 1H NMR
(DMSO-d6) 3.00 (t, N-CH2), 3.56 (t, CH2O), 3.80 (s, OCH3),
7.26 (d, Ph-H), 7.51 (d, Ph-H); 13C NMR (DMSO-d6) 50.87
(OCH3), 119.70 (C-Br), 159.91 (CdO). Anal. C, H, Br, N.
Syn th esis of 4-(4-Ch lor op h en yl)-3-m or p h olin op yr r ole-
2-ca r boxylic Acid s 23 a n d 24. Gen er a l P r oced u r e.
A
mixture of 0.03 mol of the corresponding 4-(4-chlorophenyl)-
3-morpholinopyrrole-2-carboxylic acid ester with 50 mL of
water, 50 mL of ethanol, and 20 g of NaOH was stirred and
heated to reflux for 15 min. The obtained solution was added
to water. The crystallized sodium salt of the product was
collected by filtration and then suspended in methanol. The
free 4-(4-chlorophenyl)-3-morpholinopyrrole-2-carboxylic acids
were obtained by acidification with hydrochloric acid. The
collected solid was washed with ethanol/water and dried in
vacuo.
Syn th esis of Alk yla ted 3-Mor p h olin op yr r ole-2-ca r -
boxylic Acid Meth yl Ester s (13-15). Gen er a l P r oced u r e.
A mixture of 0.05 mol of the corresponding 3-morpholinopyr-
role-2-carboxylic acid methyl ester, 0.06 mol of an alkyl halide,
45 mL CH2Cl2, 15 mL of concentrated NaOH (25%), and 0.1 g
of benzyltriethylammonium bromide was stirred at room
temperature for 8 h. The organic layer was separated, washed
with water, dried with Na2SO4, and concentrated in vacuo. The
residue was purified by recrystallization from ethanol.
The following compounds were prepared using this proce-
dure.
The following compounds were prepared using this proce-
dure.
1-Met h yl-3-m or p h olin o-4-p h en ylp yr r ole-2-ca r b oxy-
1
4-(4-Ch lor op h en yl)-3-m or p h olin op yr r ole-2-ca r b oxy-
lic a cid m eth yl ester (13): IR (KBr) 1690, 1675; H NMR
1
lic a cid (23): IR (KBr) 3240, 1660; H NMR (CDCl3) 6.85 (d,
(CDCl3) 3.84 (s, NCH3), 3.88 (s, OCH3). Anal. C, H, N.
J ) 3.42, C5H), 8.84 (s, COOH); 13C NMR (DMSO-d6) 164.47
(CdO). Anal. C, H, N.
4-(4-Ch lor op h en yl)-1-m et h yl-3-m or p h olin op yr r ole-2-
ca r boxylic a cid m eth yl ester (14): IR (KBr) 1695 cm-1; 1H
NMR (CDCl3) 3.83 (s, NCH3), 3.88 (s, OCH3). Anal. C, H, Cl,
N.
1-Ben zyl-3-m or p h olin o-4-p h en ylp yr r ole-2-ca r boxylic
a cid m eth yl ester (15): IR (KBr) 1700; 1H NMR (CDCl3) 3.82
(s, OCH3), 5.45 (s, CH2); 13C NMR (CDCl3) 50.79 2(CH3), 51.76
(NCH2), 53.22 (NCH2), 67.58 (CH2O). Anal. C, H, N.
Syn th esis of Acyla ted 3-Mor p h olin op yr r ole-2-ca r box-
ylic Acid Meth yl Ester s 16-22. Gen er a l P r oced u r e. A
solution of 0.01 mol of the corresponding 3-morpholinopyrrole-
2-carboxylic acid methyl ester in 100 mL of DMF (dry) was
stirred and cooled with ice-water. Three grams of sodium
hydride (95%) was added in small portions, and the suspension
was stirred for 15 min. The reaction mixture was filtered. The
acyl halide (0.011 mol), dissolved in 20 mL of dry DMF, was
added to the clear solution. The solution was stirred for 30
min. After addition of 200 mL of ice-water, the crystallization
started. The crude product was purified by recrystallization
from ethanol.
4-(4-Ch lor op h en yl)-1-m et h yl-3-m or p h olin op yr r ole-2-
1
ca r boxylic a cid (24): IR (KBr) 2951, 2852, 1714; H NMR
(DMSO-d6) 3.69 (s, NCH3), 13.55 (s, OH); 13C NMR (DMSO-
d6) 37.24 (CH3), 161.41 (COOH). Anal. C, H, N; Cl: calcd,
11.05; found, 11.46.
Syn th esis of 4-(4-Ch lor op h en yl)-3-m or p h olin op yr r oles
25 a n d 26. Gen er a l P r oced u r e. A mixture of 0.03 mol of
the corresponding 4-(4-chlorophenyl)-3-morpholinopyrrole-2-
carboxylic acid ester, 50 mL of water, 100 mL of ethanol, and
10 g KOH was stirred and heated to reflux for 6 h. The
resultant solution was added to 100 mL of water and extracted
with chloroform. The organic layer was washed with water,
dried with Na2SO4, and concentrated in vacuo. The residue
was purified by recrystallization from ethanol/water.
The following compounds were prepared using this proce-
dure.
4-(4-Ch lor op h en yl)-3-m or p h olin op yr r ole (25): IR (KBr)
1
3350; H NMR (CDCl3) 6.41 (t, C2-H), 6.80 (t, C5-H), 7.96 (s,
N-H); 13C NMR (CDCl3) 105.86 (C2-H). Anal. C, H, Cl, N.
The following compounds were prepared using this proce-
dure.
1-Acetyl-4-(4-flu or oph en yl)-3-m or ph olin opyr r ole-2-car -
boxylic a cid m eth yl ester (16): IR (KBr) 1725; 1H NMR
(CDCl3) 2.49 (s, CH3), 3.88 (s, OCH3). Anal. C, H, N.
4-(4-Ch lor op h e n yl)-1-m e t h yl-3-m or p h olin op yr r ole
(26): IR (KBr) 2964, 2825, 1551; 1H NMR (DMSO-d6) 3.63 (s,
NCH3), 6.40 (s, C2-H). Anal. C, H, Cl, N.
Syn th esis of 4-(4-Ch lor op h en yl)-3-m or p h olin op yr r ole-
2-a m id es (27-33). Gen er a l P r oced u r e. At a temperature
between 0 and 10 °C, 0.01 mol of thionyl chloride was added
dropwise to a stirred solution of 4-(4-chlorophenyl)-3-morpholi-
nopyrrole-2-carboxylic acid in 40 mL of dichloromethane. The
mixture was stirred under cooling with ice-water for 1 h and
then warmed to room temperature and stirred for another
hour. The hydrochloride of the acid chloride was filtered off
and dried in air with a yield of 90%. Thirty millimoles of the
amine in 10 mL of dioxane (ammonia and dimethylamine were
bubbled through the mixture) was added dropwise to a mixture
of 0.01 mol of the hydrochloride of the acid chloride in 40 mL
of dioxane at room temperature. After 2 h of stirring, the
suspension was poured into 50 g of ice. The amide was
separated and recrystallized from ethanol or methanol.
1-Acet yl-4-(4-ch lor op h en yl)-3-m or p h olin op yr r ole-2-
ca r boxylic a cid m eth yl ester (17): IR (KBr) 1725; 1H NMR
(CDCl3) 2.49 (s, CH3), 3.88 (s, OCH3); 13C NMR (CDCl3) 52.62
(OCH3), 54.50 (CH3), 150.05 (CdO), 163.79 (CdO). Anal. C,
H, Cl, N.
4-(4-Eth ylp h en yl)-3-m or p h olin o-1-p r op ion ylp yr r ole-2-
ca r boxylic a cid m eth yl ester (18): IR (KBr) 3430, 1728,
1711; 1H NMR (CDCl3) 0.98 (t, J ) 7.4, CH3), 1.23 (t, J ) 7.6,
CH3), 1.76 (m, J ) 7.4, CH2), 2.65 (m, J ) 7.6, CH2), 2.72 (m,
J ) 7.4, CH 2); 13C NMR (CDCl3) 13.60 (CH3), 15.40 (CH3),
18.20 (CH2), 28.60 (CH2), 36.60 (CH2). Anal. C, H, N.
1-Ben zoyl-4-(4-ch lor op h en yl)-3-m or p h olin op yr r ole-2-
ca r boxylic a cid m eth yl ester (19): IR (KBr) 1710, 1700;
1H NMR (CDCl3) 3.60 (s, CH3), 7.03 (s, C5H); 13C NMR (CDCl3)
51.77 (CH3), 161.89 (CdO), 167.35 (CdO). Anal. C, H, Cl,
N.
The following compounds were prepared using this proce-
dure.

Sodium Channel Blocking 3-Aminopyrroles
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 1 71
4-(4-Ch lor op h en yl)-3-m or p h olin op yr r ole-2-ca r b oxy-
lic a cid a m id e (27): IR (KBr) 3476, 3309, 1632, 1572. Anal.
C, H, N.
4-(4-Ch lor op h en yl)-3-m or p h olin op yr r ole-2-ca r b oxy-
lic a cid n -p r op yla m id e (28): IR (KBr) 3231, 1640, 1573; 1H
NMR (DMSO-d6) 0.95 (t, J ) 8.5, CH3), 1.60 (m, J ) 9.0 Hz,
CH2), 3.25 (m, J ) 8.5, CH2), 6.75 (d, C5-H) ppm. Anal. C, H,
N.
(35.3 g, 0.1 mol), was heated to reflux in 500 mL of ethanol. A
warm solution of 21 g of hydrazine hydrochloride in 25 mL of
water was added. After a few minutes a clear solution
resulted, which was heated to reflux for 2 h. The cold solution
was poured into 1.5 L of water. The precipitated solid was
collected by filtration and recrystallized from toluene to afford
14.1 g (53%) of 3-morpholino-4-(4-chlorophenyl)pyrazole, mp
1
185-188 °C; IR (KBr) 3317, 2853, 1499; H NMR (DMSO-d6)
7.88 (s, CH), 12.28 (NH); 13C NMR (DMSO-d6) 50.45 (NCH2),
66.01 (OCH2), 155.11 (C3N) ppm. Anal. (C13H14ClN3O) C, H,
Cl, N.
4-(4-Ch lor op h en yl)-3-m or p h olin op yr r ole-2-ca r b oxy-
lic a cid a llyla m id e (29): IR (KBr) 3210, 1636, 1570. Anal.
C, H, N.
4-(4-Ch lor op h en yl)-5-m or p h olin o-1,2-oxa zole (39).
A
4-(4-Ch lor op h en yl)-3-m or p h olin op yr r ole-2-ca r b oxy-
lic a cid 2-m eth oxyeth yla m id e (30): IR (KBr) 3192, 1635,
1562. Anal. C, H, N.
4-(4-Ch lor op h en yl)-3-m or p h olin op yr r ole-2-ca r b oxy-
lic a cid d im eth yla m id e (31): IR (KBr) 3198, 1618, 1574.
Anal. C, H, N.
4-(4-Ch lor op h en yl)-3-m or p h olin op yr r ole-2-ca r b oxy-
lic a cid m or p h olid e (32): IR (KBr) 3203, 1607, 1115. Anal.
C, H; N: calcd, 11.18; found, 10.56.
4-(4-Ch lor op h en yl)-3-m or p h olin op yr r ole-2-ca r b oxy-
lic a cid N-m eth ylp ip er a zid e (33): IR (KBr) 3210, 1611,
1115. Anal. C, H; N: calcd, 14.40; found, 13.86.
solution of 3.5 g (0.01 mol) of 2-(4-chlorophenyl)-3-morpholi-
nothioacrylic acid morpholide (37) and 0.69 g (0.01 mol) of
hydroxylamine hydrochloride in 20 mL of methanol was heated
to reflux for 1 h. After cooling, the precipitated solid was
collected by filtration and recrystallized from methanol to
afford 1.9
g (51%) of 4-(4-chlorophenyl)-5-morpholino-
1,2-oxazole: mp 155 °C; 1H NMR (DMSO-d6) 3.18 (t, J ) 4.5,
NCH2), 3.70 (t, J ) 4.5, OCH2), 8.55 (s, CH) ppm. Anal.
(C13H13ClN2O2) C, H, N.
1-(4-Ch lor oph en yl)-2-m or ph olin oim idazolin -4-on e (40).
This compound (mp 166-168 °C, log P 0.63) was synthesized
by Karl Gewald et al., Technical University Dresden, Institute
of Organic Chemistry (unpublished results).
2-Acetyl-4-(4-ch lor op h en yl)-3-m or p h olin op yr r ole (34).
At a temperature between 0 and 10 °C, 3.65 g (0.023 mol) of
POCl3 was added dropwise to a stirred solution of 2.07 g (0.023
mol) of N,N-dimethylacetamide in 20 mL of dichloroethane.
Then, 5.55 g (0.021 mol) of 4-(4-chlorophenyl)-3-morpholinopy-
rrole (25) dissolved in 80 mL of dichloroethane was added to
the reaction mixture. The solution was heated to reflux for 2
h. One hundred milliliters of water was added to the cold
solution. After neutralization with sodium carbonate, the
organic layer was separated, washed with water, and dried
with Na2SO4. The solution was concentrated in vacuo, and
the residue was purified by recrystallization from acetone/n-
hexane to afford 3.6 g (51%) of 2-acetyl-4-(4-chlorophenyl)-3-
morpholinopyrrole: mp 208-209 °C; 1H NMR (CDCl3) 2.63 (s,
CH3), 2.96 (t, J ) 4.7 Hz, NCH2), 3.71 (t, J ) 4.7, OCH2), 9.81
(NH); 13C NMR (CDCl3) 26.85 (CH3), 187.93 (CdO). Anal. C,
H, Cl, N.
3-[4-(4-Ch lor op h en yl)-1-m eth yl-3-m or p h olin op yr r ol-2-
yl]p r op ion ic Acid Meth yl Ester (35). A mixture of 13.85 g
(0.05 mol) of 4-(4-chlorophenyl)-1-methyl-3-morpholinopyrrole
(26), 6.45 g (0.075 mol) of acrylic acid methyl ester, 5 mL of
boron trifluoride etherate, and 200 mL of dichloromethane was
stirred for 3 h at room temperature and for 30 min at 30 °C.
Two and a half liters of water was added to the cold solution.
The organic layer was separated, washed with water, and dried
with Na2SO4. The solution was concentrated in vacuo, and
the residue was purified by recrystallization from ethanol to
yield 7 g (38%) of 3-[4-(4-chlorophenyl)-1-methyl-3-morpholi-
nopyrrol-2-yl]propionic acid methyl ester: mp 116-118 °C; 1H
NMR (DMSO-d6) 2.50 (m, CH2), 2.80 (m, CH2), 3.45 (s, NCH3),
3.65 (OCH3); 13C NMR (DMSO-d6) 19.85 (CH2), 33.52 (NCH3),
34.06 (CH2), 51.35 (OCH3), 172.38 (CdO). Anal. C, H, Cl, N.
6-(4-Ch lor oph en yl)-2-eth yl-7-m or ph olin opyr r olo[1,2-c]-
im id a zole-1,3-d ion e (36). Two hundred milligrams of ben-
zyltriethylammonium bromide and 60 mL NaOH (40%) were
added to a stirred solution of 3.3 g (0.01 mol) of 4-(4-
chlorophenyl)-3-morpholinopyrrole-2-carboxylic acid methyl
ester in 90 mL of dichloromethane. After addition of 1 mL of
ethyl isocyanate, the mixture was stirred for 30 min followed
by addition of 100 mL of water and 100 mL of dichlo-
romethane. The organic layer was separated, washed with
water, and dried with Na2SO4. The solution was concentrated
in vacuo, and the residue was purified by recrystallization from
acetonitrile to yield 2.7 g (41%) of 6-(4-chlorophenyl)-2-
ethyl-7-morpholinopyrrolo[1,2-c]imidazole-1,3-dione: mp 145-
147 °C; IR (KBr) 1775, 1715; 1H NMR (CDCl3) 1.23 (t, J )
7.1, CH3), 3.60 (q, J ) 7.1, CH2), 7.14 (s, CH); 13C NMR (CDCl3)
13.74 (CH3), 33.70 (NCH2), 147.86 (CdO), 157.48 (NCO).
Anal. (C18H18ClN2O3) C, H, Cl, N.
P h a r m a cologica l Meth od s. The evaluation for anticon-
vulsant activity was done by the Antiepileptic Drug Program,
Epilepsy Branch, Neurological Disorders Program, National
Institute of Neurological and Communicative Disorders and
Stroke. Male albino CF No. 1 mice (18-25 g, Charles River
Wilmington, MA) and male albino Sprague-Dawley rats (100
g, Simonson, Gilroy, CA) were used as experimental animals.
All animals were allowed free access to both food (S/L Custom
Lab Diet-7) and water except when they were removed from
their cages for the experiments. The test substances phenytoin
and carbamazepine were administered in 0.5% methyl cel-
lulose in water. These drugs were administered either orally
(po) or intraperitoneally (ip) in a volume of 0.01 mL/g of body
weight in mice and 0.04 mL/10 g of body weight in rats.
Deter m in a tion of Med ia n Effective Dose (ED50) or
Neu r otoxic Dose (Rotor od ). All quantitative studies were
conducted at the previously determined time of peak effect.
Groups of at least eight mice or rats were tested with various
doses of the candidate drug until at least two points were
established between the limits of 100% and 0% with respect
to protection and minimal motor impairment (neurotoxicity).
The dose of drug required to produce the desired endpoint in
50% of animals in each test and the 95% confidence interval
were then calculated by a computer program based on the
described method.50
An ticon vu lsa n t Tests. Anticonvulsant activity was es-
tablished by both electrical and chemical procedures which
have been described previously.51 One electrical test used was
the maximal electroshock seizure (MES) test. In this test, a
drop of anesthetic/electrolyte solution (0.5% butacaine hemisul-
fate in 0.9% NaCl) was applied to the eyes of each animal prior
to the placement of the corneal electrodes. The electrical
stimuli were 50 mA/60 Hz for mice and 150 mA/60 Hz for rats
delivered for 0.2 s. Abolition of the hindleg tonic extensor
component was used as the endpoint. For the kindled rat test,
the animals were daily electrically stimulated via corneal
electrodes with 8 mA/60 Hz for 4 s to a criterion of 10
consecutive stage 5 seizures, i.e. rearing and falling with
clonus.52 Abolition of the generalized seizure following sub-
stance administration was taken as the endpoint of this test.
The generalized seizure was characterized by clonus and the
rearing and falling observed in stage 4 and 5 seizures.
The chemical test used was the subcutaneous pentylene-
tetrazol (sc Met) seizure threshold test in which the convulsant
was administered to mice in doses of 85 mg/kg. In rats, a dose
of 70 mg/kg of metrazol was used. Abolition of a clonic episode
of 3 s was used as the endpoint. All subcutaneously injected
pentylenetetrazole was dissolved in 0.9% NaCl and injected
in a volume of 0.01 mL/g of body weight in rats. Mice that
4-(4-Ch lor op h en yl)-3-m or p h olin op yr a zole (38). 2-(4-
Chlorophenyl)-3-morpholinothioacrylic acid morpholide, 37

72 J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 1
Unverferth et al.
(9) Mansour, T. S.; Sauve, G. Novel Cyclization of Ethyl-2-bromopy-
ruvate with Dialkylthioamides of N-Protected Amino Acids.
Heterocycles 1988, 27, 315-318.
(10) Mencarelli, P.; Stegel, F. Different Behaviour of Secondary and
Primary Amines towards 1-Alkyl-3,4-dinitropyrroles. J . Chem.
Soc., Chem. Commun. 1980, 123-124.
(11) Gompper, R; Schneider, C. S. New (Dialkylamino)allenes,
-thiophenes and -pyrroles from Vinamidinium Salts. Synthesis
1979, 213-215.
(12) Knoll, A.; Liebscher, J . Synthesis of 3-Aminopyrrole-2-carboxylic
Acid Methyl Esters from Thioacylicamides and Glycinates.
Khim. Geterotsikl. Soedin. 1985, 628-630; Chem. Abstr. 1985,
103, 123423.
(13) Rolfs, A.; Liebscher, J . A New Synthesis of Pyrrole-2-carboxylic
acid Derivatives by Ring Transformation of 1,2-Thiazolium Salts
with Sulfur Extrusion. Angew. Chem. 1993, 105, 797-799;
Angew. Chem., Int. Ed. Engl. 1993, 32, 712-714.
(14) Rolfs, A.; Brosig, H.; Liebscher, J . Electrochemical Synthesis of
5-Amino-1,2-thiazolium Salts and Their Ring Transformation
to 3-Aminopyrroles. J . Prakt. Chem. 1995, 337, 310-312.
(15) Rolfs, A.; Liebscher, J . Efficient Synthesis of 3-Aminothioacry-
lamides by Iminoformylation of Thioacetamides. Synthesis 1994,
683-684.
(16) Liebscher, J .; Berhanu, A. Reaction of Substituted Thioamides
with Formamide Chloridessa New Synthesis of 3-Amino- and
3-Hydroxythioacrylamides. Synthesis 1982, 769-771.
(17) Brown, E. V. The Willgerodt-Reaction. Synthesis 1975, 358-
375
(18) Krall, R. L.; Penry, J . K.; White, B. G.; Kupferberg, H. J .;
Swinyard, E. A. Antiepileptic Drug Development II. Anticon-
vulsant Drug Screening. Epilepsia 1978, 19, 409-428.
(19) Lo¨scher, W.; Schmidt, D. Which Animal Models Should be Used
in the Search for New Antiepileptic Drugs? A Proposal Based
on Experimental and Clinical Considerations. Epilepsy Res.
1988, 2, 145-181.
received metrazole were observed for at least 30 min for the
presence or absence of a seizure.
Min im a l Motor Im p a ir m en t Tests (Neu r otoxicity).
Minimal motor impairment was identified in mice by the
rotorod procedure. Inability of the mouse to maintain its
equilibrium for 60 s in each of three trials on this rotating rod
was used as an indication of such impairment. In rats, motor
deficiency was determined by overt evidence of ataxia, abnor-
mal gait and stance, and/or loss of placing response and muscle
tone.
Ca lcu la tion s. The structures of all molecules were calcu-
lated at different levels of theory. For the molecular mechanics
and dynamics calculations, the QUANTA 96 program pack-
age53 was used employing the CHARMm force field.54 The
quantum chemical calculations on the semiempirical (AM1,
PM3) and ab initio (HF/3-21G) approximation levels were
performed on the basis of the SPARTAN 4.1 program pack-
age.55 All geometries were fully optimized. Systematic con-
formational searches considering all conformational degrees
of freedom were realized with AM1 and CHARMm for all
compounds to cover the complete conformational space. Most
compounds do not exhibit many conformers. In any case, the
global minimum conformations were the starting point for the
molecular dynamics simulations. Trajectories of 100 ps evolu-
tion time after heating and equilibration periods of 10 ps were
sufficient for sampling. The time steps for integration were 1
fs. The dynamics data were used for the definition of the
pharmacophore in Figure 4.
Con clu sion s
It is demonstrated that some representatives of a
series of 3-aminopyrroles which are easily accessible by
reaction of the corresponding acetophenones and glyci-
nates show high anticonvulsant activity and low neu-
rotoxicity. The structure of these compounds may well
be compared with that of other well-known and struc-
turally different anticonvulsants to suggest a general
pharmacophore for action at the voltage-dependent
sodium channel.
(20) McNamara, J . O. Development of New Pharmacological Agents
for Epilepsy: Lessons from the Kindling Model. Epilepsia 1989,
30 (Suppl.), S13-S18; discussion S64-S68.
(21) McNamara, J . O.; Byrne, M. C.; Dasheiff, R. M.; Fitz, J . G. The
Kindling Model of Epilepsy: A Review. Prog. Neurobiol. 1980,
15, 139-159.
(22) Rostock, A., Rundfeldt, C., Bartsch, R. AWD 140-190: A New
Anticonvulsant with Sodium Channel Blocking Properties. Soc.
Neurosci. Abstr. 1995, 21, 203.
(23) Reck, G. Personal communication.
(24) Scha¨fer, H. Chemical Constitution and Pharmacological Effect.
Handbook Exp. Pharmacol. 1985, 74 (Antiepileptic Drugs), 199-
243.
Ack n ow led gm en t. We thank Dr. Kupferberg and
his group at the NIH in Bethesda for their permanent
efforts and support in the investigation of 3-aminopy-
rroles. This work was supported by Bundesministerium
fu¨r Bildung und Forschung (Project “Entwicklung von
Pharmakophoren fu¨r anti-konvulsiv wirkende Verbin-
dungen”, BMBF 03C30010) and Fonds der Chemischen
Industrie.
(25) Bikker, J . A.; Kubanek, J .; Weaver, D. F. Quantum pharmaco-
logical studies applicable to the design of anticonvulsants:
Theoretical conformational analysis and structure-activity
studies of barbiturates. Epilepsia 1994, 35, 411-425.
(26) Deshmukh, T. R.; Kadaba, P. K. Identification of the triazoline
pharmacophore and the evolution of the aminoalkylpyridines,
a new class of potent orally active anticonvulsant agents. Med.
Chem. Res. 1993, 3, 223-232.
(27) Edafiogho, I. O.; Hinko, C. N.; Chang, H.; Moore, J . A.; Mulzac,
D.; Nicholson, J . M.; Scott, K. R. Synthesis and anticonvulsant
activity of enaminones. J . Med. Chem. 1992, 35, 5, 2798-2805.
(28) Scott, K. R.; Edafiogho, I. O.; Richardson, E. L.; Farrar, V. A.;
Moore, J . A.; Tietz, E. I.; Hinko, C. N.; Chang, H.; El-Assadi, A.;
Nicholson, J . M. Synthesis and anticonvulsant actvity of enami-
nones. 2. Further structure-activity correlations. J . Med. Chem.
1993, 36, 1947-1955.
(29) Edafiogho, I. O.; Alexander, M. S.; Moore, J . A.; Farrar, V. A.;
Scott, K. R. Anticonvulsant enaminones: With emphasis on
methyl 4-[(p-chlorophenyl)amino]-6-methyl-2-oxocyclohex-3-en-
1-oate (ADD 196022). Curr. Med. Chem. 1994, 1, 159-175.
(30) Camerman, A.; Camerman, N. Stereochemical similarities in
chemically different antiepilep-tic drugs. In Antiepileptic
Drugs: Mechanisms of Action; Glaser, G. H., Penry, J . K.,
Woodbury, D. M., Eds.; Raven Press: New York, 1980; pp 223-
231.
(31) Wong, M. G.; Defina, J . A.; Andrews, P. R. Conformational
analysis of clinically active anticonvulsant drugs. J . Med. Chem.
1986, 29, 562-572.
(32) J ones, G. L.; Woodbury, D. M. Principles of drug action:
structure-activity relationships and mechanisms. In Antiepi-
leptic Drugs, Woodbury, D. M.; Penry, J . K., Pippenger, C. E.,
Eds.; Raven Press: New York, 1982; pp 83-109.
(33) J ones, G. L.; Woodbury, D. M. Anticonvulsant structure-activity
relationships: Historical development and probable causes of
failure. Drug Dev. Res. 1982, 2, 333-355.
Refer en ces
(1) Perucca, E. The New Generation of Antiepileptic Drugs: Ad-
vantages and Disadvantages. Br. J . Clin. Pharmacol. 1996, 42,
531-543.
(2) Brodie, M. J .; Richens, A.; Yuen, A. W. Double Blind Comparison
of Lamotrigine and Carbamazepine in Newly Diagnosed Epi-
lepsy. Lancet 1995, 345, 476-479.
(3) Mattson, R. H. Introduction and Symposium Overview. Epilepsia
1996, 37 (Suppl. 6), S1-S3.
(4) Carson, J . R.; Carmosin, R. J .; Pitis, P. M., Vaught, J . L.; Almond,
H. R.; Stables, J . P.; Wolff, H. H.; Swinyard, E. A.; White, H. S.
Aroyl(aminoacyl)pyrroles, a New Class of Anticonvulsant Agents.
J . Med. Chem. 1997, 40, 1578-1584.
(5) Meldrum, B. S. Update on the Mechanism of Action of Antiepi-
leptic Drugs. Epilepsia 1996, 37 (Suppl. 6), S4-S11.
(6) Cirrincione, G.; Almerico, A. M.; Aiello, E.; Darralo, G. Ami-
nopyrroles. In The Chemistry of Heterocyclic Compounds; Taylor,
E. C., Ed.; J . Wiley: New York, 1992; Vol. 48, Part 2, pp 299-
523.
(7) Spitzner, R.; Schroth, W. On the 1,3-Thiazine Ring Contrac-
tion: New Aspects by σ-Donor Substitution. Tetrahedron Lett.
1985, 26, 3971-3974.
(8) de Voghel, G. J .; Eggerichs, T. L.; Clamot, B.; Viehe, H. G. Facile
Synthesis of 2-Dimethylamino-1-azirine Derivarives by Thermal-
or Photo-Isomerization of Isoxazoles. Chimia 1976, 30, 191-192.

Sodium Channel Blocking 3-Aminopyrroles
J ournal of Medicinal Chemistry, 1998, Vol. 41, No. 1 73
(34) Codding, P. W.; Duke, N. E.; Aha, L. J .; Palmer, L. Y.; McClurg,
D. K.; Szkaradzinska, M. B. Structural and computational
studies of anticonvulsants: a search for correlation between
molecular systematics and activity. In Crystallogr. Model.
Methods. Mol. Des. [Pap. Symp.] Meeting Date 1989; Bugg, E.,
Ealick, M., Eds.; Springer: New York, 1989; pp 151-160.
(35) Brouillette, W. J .; Brown, G. B.; DeLorey, T. M.; Liang, G.
Sodium channel binding and anticonvulsant activities of hydan-
toins containing conformationally constricted 5-phenyl substit-
uents. J . Pharm. Sci. 1990, 79, 871-874.
(45) Taylor, C. P.; Meldrum, B. S. Na+ channels as targets for
neuroprotective drugs. Trends Pharmacol. Sci. 1995, 16, 309-
316.
(46) Xie, X.; Lancaster, B.; Peakman, T.; Garthwaite, J . Interaction
of the antiepileptic drug lamotrigine with recombinant rat brain
type IIA Na+ channels and with native Na+ channels in rat
hippocampal neurones. Pflu¨gers Arch.sEur. J . Physiol. 1995,
430, 437-446.
(47) Stables, J . P.; Bialer, M.; J ohannessen, S. I.; Kupferberg, H. J .;
Levy, R. H.; Loiseau, P.; Perucca, E. Progress report on new
antiepileptic drugs. A Summary of the Second Eilat Conference.
Epilepsy Res. 1995, 22, 235-246.
(48) Molna´r, P.; Erdo¨, S. L. Vinpocetine is as potent as phenytoin to
block voltage-gated Na+ channels in rat cortical neurons. Eur.
J . Pharmacol. 1995, 273, 303-306.
(36) Brouillette, W. J .; Brown, G. B.; DeLorey, T. M.; Shirali, S. S.;
Grunewald, G. L. Anticonvulsant activities of phenyl-substituted
bicyclic 2,4-oxazolidenediones and monocyclic models. Compari-
son with binding to the neuronal voltage-dependent sodium
channel. J . Med. Chem. 1988, 31, 2218-2221.
(37) Brouillette, W. J .; J estkov, V. P.; Brown, M. L.; Akhtar, M. S.;
DeLorey, T. M.; Brown, G. B. Bicyclic hydantoins with
a
(49) Yamagami, Ch.; Takao, N. Hydrophobicity Parameters Deter-
mined by Reversal-Phase Liquid Chromatography. V. Relation-
ship between the Capacity Factor and the Octanol-Water
Partition Coefficient for Simple Heteroaromatic Compounds and
Their Ester Derivatives. Chem. Pharm. Bull. 1992, 40, 925-
929.
bridgehead nitrogen. Comparison of anticonvulsant activities
with binding to the neuronal voltage-dependent sodium channel.
J . Med. Chem. 1994, 37, 3289-3293.
(38) Noda, M.; Shimizu, S.; Tanabe, T.; Takai, T.; Ikeda, T.; Taka-
hashi, H.; Nakayama, H.; Kanaoka, Y.; Minamino, N.; Kangawa,
K.; Matsuo, H.; Raftery, M. A.; Hirose, T.; Inayama, S.; Ha-
yashida, H.; Miyata, T.; Numa, S. Primary structure of Electro-
phorus electricus sodium 2 channel deduced from cDNA se-
quence. Nature 1984, 312, 121-127.
(50) Finney, D. J . Probit Analysis, 3rd ed.; Cambridge University
Press: London, 1971.
(51) Swinyard, E. A.; Woodhead, J . H.; White, H. S.; Franklin, M. R.
General Principles: Experimental Selection, Quantification, and
Evaluation of Anticonvulsants. In Antiepileptic Drugs, 3rd ed.;
Levy, R. H., Dreifuss, F. E., Mattson, R. H., Meldrum, B. S.,
Penry, K. J ., Eds.; Raven Press: New York, 1989; pp 85-102.
(52) Racine, R. J . Modification of Seizure Activity by Electric
Stimulation: II. Motore Seizure. Electroenceph. Clin. Neuro-
physiol. 1972, 32, 281-294.
(39) Noda, M.; Ikeda, T.; Kayano, T.; Suzuki, H.; Takeshima, H.;
Kurasaki, M.; Takahashi, H.; Numa, S. Existence of distinct
sodium channel messenger RNAs in rat brain. Nature 1986, 320,
188-192.
(40) Trimmer, J . S.; Cooperman, S. S.; Tomiko, S. A.; Zhou, J .; Crean,
S. M.; Boyle, M. B.; Kallen, R. G.; Sheng, Z.; Barchi, R. L.;
Sigworth, F. J .; Goodman, R. H.; Agnew, W. S.; Mandel, G.
Primary structure and functional expression of a mammalian
skeletal muscle sodium channel. Neuron 1988, 3, 33-49.
(41) Salkoff, L.; Butler, A.; Wie, A.; Scavarda, N.; Giffen, K.; Ifune,
C.; Goodman, R.; Mandel, G. Genomic organization and deduced
amino acid sequence of a putative sodium channel gene in
Drosophila. Science 1987, 237, 744-748.
(53) QUANTA 96 Molecular Modelling Program Package, Molecular
Simulations, Inc., San Diego, 1996.
(54) Momany, F. A.; Rone, R. Validation of the General Purpose
QUANTA 3.2/CHARMm force field. J . Comput. Chem. 1992, 13,
888-900.
(55) SPARTAN, version 4.1, Wavefunction, Inc., Irvine, 1994.
(56) Faigle J . W.; Menge G. P. Pharmacokinetic and metabolic
features of oxcarbazepine and their clinical significance: com-
parison with carbamazepine. Int. J . Clin. Psychopharmacol.
1990, 3 (Suppl. 1), 73-82.
(42) Catterall, W. A. Structure and function of voltage-sensitive ion
channels. Science 1988, 242, 50-61.
(43) Catterall, W. A. Common modes of drug action on Na+ chan-
nels: Local anesthetics, antiarrhythmics and anticonvulsants.
Trends Pharmacol. Sci. 1987, 8, 57-65.
(44) Willow, M.; Catterall, W. A. Inhibition of binding of [3H]-
Batrachotoxinin A 20-R-benzoate to sodium channels by the
anticonvulsant drugs diphenylhydantoin and carbamazepine.
Mol. Pharmacol. 1982, 22, 627-635.
J M970327J