(Aryloxy)aryl semicarbazones and related compounds: A novel class of anticonvulsant agents possessing high activity in the maximal electroshock screen

Article abstract of DOI:10.1021/jm9603025

A number of (aryloxy)aryl semicarbazones and related compounds were synthesized and evaluated far anticonvulsant activities. After intraperitoneal injection to mice, the semicarbazones were examined in the maximal electroshock (MES), subcutaneous pentylenetetrazole (scPTZ), and neurotoxicity (NT) screens. The results indicated that greater protection was obtained in the MES test than the scPTZ screen. Quantitation of approximately one-third of the compounds revealed an average protection index (PI, i.e. TD₅₀/ED₅₀) of approximately 9. After oral administration to rats, a number of compounds displayed significant potencies in the MES screen (ED₅₀ of 1-5 mg/kg) accompanied by very high protection indices. In fact over half the compounds had PI figures of greater than 100, and two were in excess of 300. The compounds were essentially inactive in the scPTZ and NT screens after oral administration to rats. Various compounds displayed greater potencies and PI figures in the mouse intraperitoneal and rat oral screens than three reference clinically used drugs. The data generated supported a binding site hypothesis. Quantitative structure-activity relationships indicated a number of physicochemical parameters which contributed to activity in the MES screen. X-ray crystallography of five compounds suggested the importance of certain interatomic distances and bond angles for activity in the mouse and rat MES screens.

Full text of DOI:10.1021/jm9603025

3984
J . Med. Chem. 1996, 39, 3984-3997
(Ar yloxy)a r yl Sem ica r ba zon es a n d Rela ted Com p ou n d s: A Novel Cla ss of
An ticon vu lsa n t Agen ts P ossessin g High Activity in th e Ma xim a l Electr osh ock
Scr een
J onathan R. Dimmock,*,† Ramanan N. Puthucode,† J ennifer M. Smith,† Mark Hetherington,† J . Wilson Quail,‡
Uma Pugazhenthi,‡ Terry Lechler,‡ and J ames P. Stables§
College of Pharmacy and Nutrition and Department of Chemistry, University of Saskatchewan,
Saskatoon, Saskatchewan, S7N 5C9 Canada, and National Institute of Neurological Disorders and Stroke,
Bethesda, Maryland 20892-9020
Received April 23, 1996X
A number of (aryloxy)aryl semicarbazones and related compounds were synthesized and
evaluated for anticonvulsant activities. After intraperitoneal injection to mice, the semicar-
bazones were examined in the maximal electroshock (MES), subcutaneous pentylenetetrazole
(scPTZ), and neurotoxicity (NT) screens. The results indicated that greater protection was
obtained in the MES test than the scPTZ screen. Quantitation of approximately one-third of
the compounds revealed an average protection index (PI, i.e. TD50/ED50) of approximately 9.
After oral administration to rats, a number of compounds displayed significant potencies in
the MES screen (ED50 of 1-5 mg/kg) accompanied by very high protection indices. In fact
over half the compounds had PI figures of greater than 100, and two were in excess of 300.
The compounds were essentially inactive in the scPTZ and NT screens after oral administration
to rats. Various compounds displayed greater potencies and PI figures in the mouse
intraperitoneal and rat oral screens than three reference clinically used drugs. The data
generated supported a binding site hypothesis. Quantitative structure-activity relationships
indicated a number of physicochemical parameters which contributed to activity in the MES
screen. X-ray crystallography of five compounds suggested the importance of certain interatomic
distances and bond angles for activity in the mouse and rat MES screens.
In tr od u ction
A previous study revealed that a number of aryl
semicarbazones possessed anticonvulsant activity in the
maximal electroshock (MES) and subcutaneous penty-
lenetetrazole (scPTZ) screens when administered by the
intraperitoneal route to mice.1 These two test systems
have been claimed to detect compounds affording pro-
tection to generalized tonic-clonic seizures and general-
ized absence convulsions respectively.2 Nevertheless
the compounds displayed neurotoxicity when given by
this route and the protection indices (PI viz. TD50/ED50
where TD50 and ED50 refer to the doses eliciting neu-
rotoxicity and anticonvulsant activities respectively in
F igu r e 1. Proposed binding site of aryl semicarbazones.
50% of the animals) of 10 representative compounds
were low, i.e. in general less than 4. However an oral
administration to rats, two interesting features were
observed. First, marked activity in the MES screen was
at complementary areas on a macromolecular complex
in vivo; these areas have been referred to as the
hydrogen bonding area and the aryl binding site,
respectively.4 These possible interactions are repre-
sented in Figure 1. The principal aim of the present
study was to investigate the area around the postulated
aryl binding site, which is shown in Figure 1 as the
auxiliary binding area. Additional groups placed on the
aryl ring could strengthen attachment at the binding
site and increase potency, or alternatively, steric inter-
actions, for example, may exert a dystherapeutic effect.
From these investigations a clearer picture of the nature
of the postulated binding site may emerge. A second
goal of this investigation was the preparation of orally
active compounds with approximately 10-fold greater
potency than the semicarbazones synthesized previ-
ously, i.e. they should possess ED50 figures in the 2-3
mg/kg range while still retaining favorable PI values.
If this objective was realized, structure-activity rela-
noted whereby some of the compounds had ED50 figures
in the 20-25 mg/kg range while activity in the scPTZ
test was virtually abolished. Second, neurotoxicity was
diminished, and PI figures of approximately 25 were
detected in some of the compounds.1 A subsequent
study of the anticonvulsant activities of a number of
related aryl semicarbazones confirmed these general
trends.3
If the aryl semicarbazones displaying activity in the
MES screen interact at a specific binding site, it is likely
that the semicarbazono group and the aryl ring align
* To whom inquiries should be directed.
† College of Pharmacy and Nutrition, Saskatoon.
‡ Department of Chemistry, Saskatoon.
§ National Institute of Neurological Disorders and Stroke, Bethesda.
X Abstract published in Advance ACS Abstracts, August 15, 1996.
S0022-2623(96)00302-0 CCC: $12.00 © 1996 American Chemical Society

(Aryloxy)aryl Semicarbazones and Related Compounds
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 20 3985
tionships in this new group of potent anticonvulsants
would be sought using various physicochemical and
computational techniques.
Table 1 reveals that the semicarbazone 58 was
inactive in the mouse intraperitoneal screen and af-
forded only weak protection when given orally to rats
in the MES test. Compounds 59-66 were proposed
since reduction or abolition of anticonvulsant properties
in these analogs would confirm the limitations as to the
places on the auxiliary binding site at which interactions
with the substituted aryl ring occurs. The semicarba-
zones prepared previously had an oxygen atom between
the two aryl rings. Compounds 67-72 were designed
that incorporated different spacer groups which could
affect not only the distances between the two aryl rings
but also their orientation in relation to each other. In
other words, a lack of coplanarity between the proximal
and distal rings may vary among the compounds 67-
72, and hence the biodata generated would afford some
insight into the structure of the distal binding site.
Likewise the preparation of 73-83 was considered
whereby isosteric replacement of the oxygen atom of
representative semicarbazones among the compounds
2-57 by sulfur was planned. In order to gain an
improved understanding of the steric and electronic
requirements of the hydrogen bonding area, the decision
was made to prepare compounds 84-92 whose anticon-
vulsant activity could be compared to certain oxo
analogs in the semicarbazones 2-48 and 73-83. The
R1 groups in 84-92 would have sizes and hydrogen-
bonding capabilities different than the oxo and amino
groups found in the analogous semicarbazones. Fur-
thermore, the question of whether replacement of the
phenyl ring of 2 by â-naphthyl and 4-pyridyl groups
producing compounds 93 and 94 would lead to com-
pounds which would align at the distal binding site was
posited. Finally, the possibility exists that by having
two semicarbazono and two phenoxyaryl groups in one
molecule a significant increase in anticonvulsant prop-
erties may be obtained since two rather than one parts
of the molecule may align at the binding site. On the
other hand, should 95 display little or no anticonvulsant
activity, the result may indicate steric impendance at
the binding site, i.e. the molecule has too large a group
at the 4′-position of a phenoxyaryl group to be accom-
modated.
Initially the attachment of a second aryl group,
designated the distal ring, to the proximal aromatic ring
located nearest to the semicarbazono group was con-
sidered, i.e. the preparation and anticonvulsant evalu-
ation of 4-phenylbenzaldehyde semicarbazone. This
compound could increase the van der Waals bonding at
a binding site and increase potency. However it dis-
played no bioactivity in the MES, scPTZ, and neuro-
toxicity screens when doses up to and including 300 mg/
kg were given intraperitoneally to mice.3 Nevertheless
oral administration to rats revealed that neurotoxicity
was absent at a dose of 500 mg/kg and activity in the
MES screen was displayed, although the results were
ambiguous, i.e. 0/8, 2/8, 3/8, and 0/16 animals were
protected using doses of 20, 40, 80, and 160 mg/kg
respectively. No oral activity in rats was noted in the
scPTZ screen at a dose of 250 mg/kg. The conclusion
reached was that the presence of a distal aromatic ring
did not induce neurotoxicity and MES activity, while
weak, was retained when the compound was adminis-
tered orally. Thus the realignment of the distal ring at
a distance from the proximal ring greater than is found
with this compound was considered, i.e. the placement
of a spacer group between the two aryl rings.
These considerations led to the decision to synthesize
the three structural isomers 1, 2, and 58, whereby a
phenoxy group was placed in different locations of the
proximal aryl ring, i.e. an oxygen atom was used as the
spacer group. The structures of the compounds may be
ascertained from Scheme 1 and Table 1. While the
ortho (58) and meta (1) isomers were either inactive or
demonstrated weak anticonvulsant properties in the
mouse and rat screens (Table 1), the para analog 2
afforded good protection in both tests. From this
observation, molecular modification of 2 proceeded in
several directions with a view to defining more clearly
the nature of the binding site for these compounds.
It is conceivable that while the phenoxy groups can
align at different places on the auxiliary binding site,
the presence of a distal ring in the location “para” to
the proximal aromatic ring is preferable. In other
words, favorable interaction of the distal ring and an
area designated the distal binding site may occur
(Figure 2). Early in this study, the marked activities
of 5 and 23 were observed; hence the aryl substitution
pattern in this series was developed using principally
halogens and in particular fluorine (3-20) as well as
alkyl groups (21-38). These aryl substituents were also
used in other series of compounds in order that com-
parisons of their bioactivities could be made with these
semicarbazones. The preparation and bioevaluation of
the semicarbazones 3-48 may permit discernment of
the structural and electronic requirements at the distal
binding site. In addition, replacement of the methine
proton by small alkyl groups leading to 49-57 was
considered since the alkyl groups could interact at a
hydrophobic pocket by forming van der Waals bonds,
thereby assisting alignment at the binding site and
leading to an increase in anticonvulsant activity. Al-
ternatively unfavorable steric interactions between the
alkyl groups in 49-57 and a place on the binding site
could occur, resulting in a lowering of potency.
In summary, the principal aim of this study was to
prepare a number of (aryloxy)aryl semicarbazones and
related compounds for evaluation as candidate anticon-
vulsants with a view to understanding their chemical
features which contribute to interactions at a binding
site.
Resu lts
The compounds were synthesized using the method-
ologies outlined in Scheme 1.
All of the compounds were examined in the MES,
scPTZ, and NT screens after intraperitoneal injection
into mice and these data are presented in Table 1.
Quantitation of some of the compounds was undertaken,
and the results are summarized in Table 2. After oral
administration to rats, various semicarbazones and
related compounds were evaluated for activity princi-
pally in the MES screen, and the results are given in
Tables 1 and 3. X-ray crystallography of 1, 2, 5, 23,
and 73 was undertaken in order to obtain further
information pertaining to the structural features of a
putative binding site.

3986 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 20
Dimmock et al.
Sch em e 1. Preparation of (Aryloxy)aryl Semicarbazones and Related Compounds 1-95a
O
O
H
N
a
C
O
N
H2N
O
C
C
H
H
1
R1
R2
R1
R2
H
N
O
R1
R2
F
F
O
O
a
O
O
+
+
C
O
N
N
C
C
HO
HO
H2N
O
C
C
C
C
H
H
H
2-48
H
N
R1
O
R1
a
R1
O
C
O
H2N
R2
R2
R2
49-57
R1
R1
O
O
F
H
N
a
C
O
N
H2N
C
HO
R1
O
+
C
R2
O
C
R2
R2
58-66
OCH2
a
O
C
H
H
N
R2
R1
R1
R2
OH
a
O
C
O
N
H2N
O
O
O
C
+
C
C
C
C
CIX
R2
H
H
H
H
67-72
F
SO2
O
+
+
C
NaSO2
a
H
H
N
R1
S
R1
F
S
a
R1
O
O
C
O
N
H2N
C
C
C
HS
R2
R2
R2
73-83
H
N
R2
F
R2
F
F
b, c
C
HR2
F
C
N
H2N
O
C
+
H
H
H
R1
84-87
d
H
N
R2
F
R1
C
N
C
H
88-92
O

(Aryloxy)aryl Semicarbazones and Related Compounds
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 20 3987
H
OR1
OR1
a
C
O
N
N
H2N
O
C
C
R2
93-94
R2
O
O
F
+
HO
O
OH
C
O
C
O
O
2
C
H
H
a
R2
H
N
H
N
O
C
N
H2N
N
C
O
C
C
NH2
H
H
O
95
a
a ) semicarbazide, b ) thiosemicarbazide, c ) aminoguanidine, d ) various hydrazides possessing the general formula RCONHNH2,
X ) CO or SO2.
Discu ssion
83); 100, 100, 100 (84-87); and 80, 80, 100 (88-92).
These figures indicate that, in the MES screen, the per-
centage of active compounds was similar to 2-48 except
in the case of the group of semicarbazones 58-66. The
data for the scPTZ test showed that in comparison to
2-48, activity was greater in the series 49-57, 73-83,
84-87, and 88-92 and lower in the analogs 58-66 and
67-72. In contrast to 2-48, neurotoxicity was more
widely detected in the series 49-57, 84-87, and 88-
92; similar in 73-83; and lower in 58-66 and 67-72.
Thus, in general, activity was retained by molecular
modification of 2-48 except for 58-66, suggesting that
placement of aryloxy groups in the para position of the
proximal ring is preferable to the ortho position.
A major aim of this study was to evaluate the viability
of the binding site hypothesis (Figure 2). The method
followed involved first the synthesis and anticonvulsant
examination of different series of aryloxyaryl semicar-
bazones and related compounds. Second, an evaluation
was made of the data using structure-activity relation-
ships (from both a qualitative and quantitative view-
point) and X-ray crystallography.
The syntheses of compounds 1-95 were accomplished
successfully. The anticonvulsant examination of these
compounds in the mouse intraperitoneal screen will be
reviewed initially; subsequently the evaluation of most
of these compounds in the rat oral test will be discussed.
The data in Table 1 indicate the evaluation of all of
these molecules in the MES, scPTZ, and NT screens
after intraperitoneal injection in mice using doses of 30,
100, and 300 mg/kg. The following observations may
be made. First, in general, the compounds demon-
strated a selective protection in the MES screen rather
than the scPTZ test which may be noted by comparing
not only the percentage of compounds which were active
in both screens but the fact that lower doses were
required to afford protection. Thus the percentage of
compounds which were active at minimum doses of 30,
100, and 300 mg/kg or were inactive were 56, 18, 15,
and 11, respectively, in the MES screen whereas in the
scPTZ test, the comparable figures were 16, 18, 25 and
41 respectively. Second, the percentage of compounds
causing neurological deficit at minimum doses of 30,
100, and 300 mg/kg or were not neurotoxic at the
maximum dose utilized were 5, 30, 35, and 30 respec-
tively. Thus in the 30-100 mg/kg dose range, while
65% of the compounds did not display neurotoxicity,
74% demonstrated activity in the MES screen which
suggested that favorable PI values may be displayed in
this group of compounds. Third, a comparison of the
bioactivity found in compounds 2-48 with the groups
of analogs in which four or more compounds were
present was made in order to discern the general effects
of structural modifications of the series comprising the
semicarbazones 2-48 on anticonvulsant activity. The
percentage of compounds in this series which dis-
played activity in the MES, scPTZ, and NT screens was
96, 58, and 75, respectively. The comparable figures for
the analogs were as follows: 100, 78, 89 (49-57); 33,
22, 56 (58-66); 83, 33, 17 (67-72); 100, 82, 82 (73-
Quantitative evaluations of the anticonvulsant ef-
ficacy and neurotoxicity of approximately one-third of
the compounds were undertaken, and the results are
presented in Table 2. Some of the features of interest
are as follows. First, the selective efficacy of these
anticonvulsants in reducing seizures in the MES rather
than the scPTZ screen was confirmed. Thus all of the
compounds displayed activity in the MES test, whereas
ED50 figures were obtained in only 38% of these analogs
for the scPTZ screen. In the remaining cases, the dose
was elevated substantially above the ED50 figure in the
MES screen. Furthermore, a comparison of the PI
values in 12 cases where ED50 figures were available
in both the MES and scPTZ screens revealed that higher
figures were always obtained from MES screening. In
fact the average PI value of the MES screen was 5.2
times the ED50 figures generated in the scPTZ test. The
PI values of 32 and 82 which are approximately 22 and
21, respectively, are particularly noteworthy. Subse-
quent discussion will therefore revolve around the MES
screening. Second, a comparison was made with the
three reference drugs phenytoin, carbamazepine, and
valproate in terms of potency and PI values. The ED50
figures in the mouse MES screen for compounds 18, 25,
52, 57, and 81 were lower than for phenytoin and
carbamazepine; in addition those of 35, 49, 53, 56, and
82 were less than that of carbamazepine. Furthermore,
the percentage of compounds which had PI values
greater than phenytoin and carbamazepine were 63 and
78, respectively. All of the compounds listed in Table 2
were considerably more potent and had PI values higher
than that of valproate. Hence a number of these
(aryloxy)aryl semicarbazones and related compounds

3988 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 20
Dimmock et al.
Ta ble 1. Aryl Substituents, Physical Data, and Anticonvulsant Evaluation after Intraperitoneal Injection into Mice and Oral
Administration to Rats of the Compounds 1-95
oral administration to rats:b
intraperitoneal injection in micea
MES screen
aryl substituents
R1 R2
---
MES screen scPTZ screen toxicity screen
yield
(%)
dose
(mg/kg) 0.25 h 0.5 h 1 h 2 h 4 h
compd
mp (°C)
0.5 h
4 h
0.5 h
4 h
0.5 h
4 h
1
2
3
4
5
6
7
8
9
---
224-225
224-225
228-230
209
233-234
225
229-230
230
232
212-213
177
207-208
185-186
225-226
216-217
225-226
221-222
225-226
209-210
203-205
205
70
60
42
42
65
50
42
65
27
86
46
42
35
40
45
60
71
60
59
40
25
35
50
52
63
70
25
60
62
40
48
53
63
38
48
61
40
30
60
41
63
58
72
60
35
20
55
40
60
58
74
72
60
38
30
38
62
40
14
48
27
55
32
15
28
5
-
100
100
30
30
100
30
100
30
100
30
30
30
30
300
30
30
30
30
100
30
30
30
300
300
300
300
100
100
30
300
30
30
30
30
100
30
30
30
30
30
30
100
100
100
100
100
30
100
100
30
-
-
300
100
-
300
100
100
300
30
-
-
-
-
-
50
-
-
2
-
3
-
4
4
4
4
4
4
1
4
4
4
4
4
4
4
4
4
4
4
4
3
4
4
1
4
1
4
4
3
4
4
2
-
2
4
2
x
x
4
4
4
4
-
4
-
x
-
4
4
3
4
4
4
4
4
4
2
2
x
1
3
-
x
x
x
x
1
-
1
x
2
x
-
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
4
3
4
-
4
-
4
4
2
4
4
4
1
3
4
3
x
x
4
4
4
4
3
4
-
x
1
4
4
3
4
4
4
4
4
4
4
1
x
2
3
-
x
x
x
x
1
-
-
x
4
x
-
4
4
4
4
4
4
1
4
4
4
4
4
4
4
4
4
4
4
4
4
2
4
1
4
-
4
4
-
4
4
2
2
4
4
4
x
x
4
4
4
4
1
4
2
x
1
4
4
-
4
4
4
4
4
4
4
1
x
1
2
-
x
x
x
x
-
-
2
x
3
x
H
H
-
-
-
-
50
2-F
3-F
4-F
2-F
2-F
2-F
2-F
3-F
3-F
2-Cl
3-Cl
4-Cl
3-Cl
H
H
H
3-F
4-F
5-F
6-F
4-F
5-F
H
-
50
4
-
300
-
-
-
300
300
-
300
300
300
300
-
300
-
-
-
300
300
-
300
-
100
30
50
4
4
-
50
2
4
-
12.5
50
-
3
3
4
-
-
12.5
12.5
50
-
-
2
1
2
4
300
300
-
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
47
48
49
50
51
52
53
54
55
56
57
58
59
60
61
62
63
64
65
66
67
68
69
70
71
72
12.5
50
1
3
3
4
100
30
30
-
300
300
-
H
H
4-Cl
H
50
50
-
4
4
4
-
-
300
30
100
30
50
50
-
1
2
4
4-Br
-
300
300
100
100
300
300
100
-
4-I
H
100
-
300
-
50
3
4
2-F
2-Cl
2-Br
4-Cl
4-F
4-F
H
H
H
12.5
50
2
4
4
4
-
-
-
300
300
300
300
-
300
100
-
50
4
4
2-CH3
3-CH3
4-CH3
2-CH3
2-CH3
2-CH3
2-CH3
3-CH3
3-CH3
4-C2H5
2-C3H7
4-C3H7
4-C4H9
4-C4H9
4-C4H9
4-C5H11
4-C8H17
4-C8H17
2-CH3
2-CH3
3-CH3
3-CH3
4-C6H5
4-OCH3
4-OC4H9
4-OC7H15
100
-
12.5
12.5
50
-
-
3
4
4
4
205-206
219-221
-
-
-
-
3-CH3 222-223
4-CH3 193-195
5-CH3 210-211
6-CH3 190
4-CH3 216-218
5-CH3 209-210
H
H
H
H
H
H
H
H
-
30
100
30
100
30
-
-
12.5
12.5
12.5
30
-
-
30
30
30
300
300
300
-
30
300
100
-
300
-
100
300
300
300
-
1
4
1
1
-
4
300
-
12.5
12.5
12.5
12.5
12.5
12.5
12.5
12.5
15
-
-
-
3
-
-
-
-
-
x
2
4
2
3
1
-
2
-
-
x
300
30
30
100
100
30
30
3
300
-
30
-
-
210
175-177
215
30
-
300
300
-
100
-
n
n
n
s
100
100
300
100
100
300
-
-
30
30
30
30
-
100
300
-
-
30
30
30
30
30
30
30
30
30
30
-
300
-
-
-
-
300
-
-
300
300
100
100
-
215
-
-
192-193
200-202
198-200
218-220
190
196-197
156-157
224
220
280
218-220
203
204-206
209-210
218-220
169-171
154-156
182-184
170-172
192-194
186-188
195-197
184-186
205
198-199
136-138
210-212
154-157
156-158
167-169
136-138
183-186
155-157
212-213
237-238
245-246
254
-
100
100
-
300
100
-
t
-
t
-
n
t
-
-
-
-
-
-
x
H
-
300
100
100
-
x
x
x
4-Cl
4-I
4-F
4-Cl
H
H
H
H
30
-
30
-
300
300
300
300
-
12.5
12.5
12.5
12.5
12.5
50
-
3
4
4
4
-
4
-
x
30
30
30
2
-
-
-
2
1
-
-
-
x
300
-
100
300
300
300
-
300
100
100
-
300
30
100
100
30
30
30
-
30
300
-
-
-
n
300
-
300
-
300
300
-
12.5
x
n
4-OC6H5
4-CN
H
H
F
F
Cl
Cl
Br
H
H
-
-
-
50
12.5
30
-
-
30
-
30
-
300
100
100
300
300
30
100
100
100
100
100
100
100
100
100
-
2
4
CH3
C2H5
CH3
C2H5
CH3
C2H5
CH3
C2H5
CH3
H
4
4
-
-
30
1
4
100
100
-
-
12.5
12.5
30
-
-
3
4
2
4
-
30
-
300
300
100
30
-
300
300
300
-
30
-
1
3
2
1
-
x
-
12.5
12.5
12.5
50
1
Br
CH3
H
30
-
-
-
-
x
100
-
-
-
-
-
H
F
CH3
H
-
300
-
-
300
-
-
x
-
-
-
30
30
-
1
-
1
F
F
Cl
Cl
Br
Br
OCH2
OCO
OCO
SO2
OSO2
OSO2
CH3
C2H5
CH3
C2H5
CH3
C2H5
H
H
Cl
H
H
-
-
-
-
-
-
-
-
-
300
300
-
12.5
x
-
x
-
x
300
300
-
300
300
-
300
-
300
-
300
-
x
x
x
-
-
-
300
-
-
x
x
x
-
-
-
300
-
x
x
x
52
70
80
40
40
70
300
-
300
300
300
300
30
-
100
-
-
12.5
12.5
12.5
x
-
-
1
-
-
-
x
-
-
-
-
-
-
-
-
-
-
-
-
-
x
146
205-207
30
30
-
-
300
-
300
-
12.5
x
1
x
2
x
CH3
-
-
-

(Aryloxy)aryl Semicarbazones and Related Compounds
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 20 3989
Ta ble 1 (Continued)
oral administration to rats:b
intraperitoneal injection in micea
MES scrren scPTZ screen toxicity screen
MES screen
aryl substituents
R1 R2
yield
(%)
dose
(mg/kg) 0.25 h 0.5 h 1 h 2 h 4 h
compd
mp (°C)
0.5 h
4 h
0.5 h
4 h
0.5 h
4 h
73
74
75
76
77
78
79
80
81
82
83
84
85
86
87
88
89
90
91
92
93
94
95
H
H
H
F
F
F
Cl
Br
Br
CH3
CH3
S
H
226-227
208-210
131-133
230-231
204-207
150-152
216
212-213
214-216
225-227
222-224
156-158
171-172
181-183
172-173
176-178
146-148
160
40
60
16
52
91
18
40
30
46
32
60
56
62
50
40
60
80
83
80
75
61
15
50
30
100
30
30
100
30
100
100
100
30
100
30
100
300
300
300
100
30
300
-
100
-
-
30
30
-
30
100
30
30
30
100
30
30
30
30
-
300
100
30
-
-
300
-
300
100
100
30
100
30
-
-
-
300
-
50
30
30
12.5
30
-
-
-
1
3
-
1
4
4
3
3
4
-
4
1
3
4
4
4
4
4
2
4
3
4
4
4
3
4
4
3
4
4
4
4
3
3
1
1
1
x
-
2
-
x
4
-
x
x
x
x
4
4
4
4
4
3
4
4
4
4
4
1
1
2
-
x
1
1
-
x
4
2
x
x
x
x
CH3
C2H5
H
CH3
C2H5
H
H
CH3
H
CH3
O
S
O
S
O
S
O
O
O
H
H
---
-
100
-
100
300
300
100
-
100
100
300
100
100
300
300
100
-
300
-
30
50
-
300
-
-
-
12.5
12.5
12.5
12.5
12.5
12.5
12.5
12.5
x
-
1
100
-
300
-
-
-
-
-
-
-
x
1
1
-
x
1
-
x
x
x
x
-
-
2
4
-
2
100
30
100
30
30
100
-
30
S
100
-
30
100
100
100
300
300
100
300
300
-
-
-
100
300
-
1
2
NH
NH
H
100
100
-
-
-
x
-
4
-
x
-
-
x
-
-
x
-
2
-
x
3
-
x
-
30
-
-
300
100
30
300
300
100
300
-
H
-
30
12.5
30
CH3
H2NNH
H2NCO
100
-
100
-
220
253
240-242
300
100
-
100
300
-
30
100
-
-
300
-
x
c
C
10H7
-
300
-
12.5
12.5
x
x
x
C5H4Nd
---
-
-
-
342-344
-
-
-
phenytoin
carbamazepine
valproic acid
-
100
100
-
x
x
x
x
x
x
100
300
300
-
x
a
Doses of 30, 100, and 300 mg/kg were administered. The figures in the table indicate the minimum dose whereby bioactivity was
demonstrated in half or more of the mice. The animals were examined 0.5 and 4 h after injections were made. The lines - indicate an
b
absence of anticonvulsant activity and neurotoxicity at the maximum dose administered (300 mg/kg). The figures in the screen indicate
the number of rats out of four which were protected. The line - means that no activity was demonstrated and the designation × indicates
d
that the compound was not screened. c â-Naphthyl group. 4-Pyridinyl group.
the average PI values for the compounds in series 2-48,
49-57 (based on seven of the eight compounds for which
both ED50 and TD50 figures were available), and 73-
83 were 8.95, 7.80, and 9.85, respectively, indicating
that neurotoxicity was lower in series 73-83. The data
recorded in Table 2 do not conflict with the hypothesis
that these compounds align at the binding site repre-
sented in Figure 2.
In order to evaluate the effects of different substitu-
ents in the distal aryl ring on anticonvulsant activity
and neurotoxicity, various linear and semilogarithmic
plots were made. The relatively few ED50 figures in the
scPTZ test precluded their consideration; hence com-
ments will be confined to activities in the MES screen.
The physicochemical constants chosen reflected the
electronic (σ,σ*), hydrophobic (π), steric (MR, i.e. molar
F igu r e 2. Proposed binding site of (aryloxy)aryl semicarba-
refractivity), and topological (SA, i.e. surface area)
zones.
characteristics of the aryl substituents. A correlation
coefficient of 0.8 was chosen arbitrarily as indicating a
relationship between the physicochemical constants and
bioactivity, and when observed, the specific r values are
indicated in the Experimental Section.
compare favorably with these three reference drugs.
Third, with the exception of the data for 71, the
compounds listed in Table 2 are found in three series
of compounds, namely 2-48, 49-57, and 73-83. Since
Plots were made between the σ,σ*, π, MR, and SA
constants of the aryl substituents in the distal ring
against the MES ED50 figures of (i) all members of series
2-48 listed in Table 2; (ii) 49, 51, 53, 55, and 57; (iii)
52, 54, and 56; and (iv) 73, 76, 79, 80, and 82. In
addition, in order to obtain some insight into the
importance of the size and shape of the group attached
to the azomethine carbon atom and also at the para
position of the distal aryl ring, the ED50 figures of (v) 5,
51, and 52 and (vi) 5, 23, 30, 32, 34, and 35 were plotted
against the MR and SA values.
the substituents in the aryl ring were not constant in
each of these three series, a strict comparison between
the potencies and PI values cannot be made. However
the following general observations may be of value in
subsequent drug design. The average ED50 figures in
the MES screen for the compounds in series 2-48, 49-
57, and 73-83 were 15.90, 8.90, and 13.64 mg/kg,
respectively, i.e. the highest potency was found in series
49-57. It is conceivable therefore that replacement of
the methine proton by alkyl groups may indicate a
hydrophobic bonding area on the binding site. However

3990 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 20
Dimmock et al.
Ta ble 2. Evaluation of Selected Compounds in the MES, scPTZ, and Neurotoxicity Screens after Intraperitoneal Injection in Mice
MES screen
scPTZ screen
neurotoxicity screen
PIa
ED50 (mg/kg)
(95% CI)
slope
(SE)
ED50 (mg/kg)
(95% CI)
slope
(SE)
TD50 (mg/kg)
(95% CI)
slope
(SE)
compd
t (h)
t (h)
t (h)
MES
8.22
scPTZ
3
0.5
20.7
(18.7-22.1)
12.9
(10.5-17.1)
45.8
(41.4-52.2)
11.4
(6.68-19.2)
11.3
(8.31-12.9)
14.5
(9.53-18.9)
12.3
(8.24-18.4)
27.7
(20.4-36.1)
3.31
(2.23-4.17)
13.1
(8.70-20.1)
14.7
(10.4-19.2)
4.60
(3.27-5.75)
34.7
(27.1-43.0)
10.3
(8.90-11.4)
11.0
(9.27-12.7)
13.4
(10.4-16.3)
8.87
(7.70-9.96)
9.08
(6.45-11.3)
11.6
(11.0-12.5)
5.46
(4.57-6.46)
9.14
(7.42-11.1)
11.1
(10.4-12.6)
12.5
(9.47-15.2)
6.71
(5.56-7.98)
5.62
(3.67-8.30)
25.3
(21.5-29.9)
15.6
(10.5-20.6)
12.4
(9.25-16.1)
16.2
(14.6-17.6)
24.4
(18.5-30.9)
3.82
(2.91-4.84)
9.46
18.6
(5.63)
8.28
(3.00)
15.5
(5.71)
2.78
(0.86)
10.9
(4.27)
4.62
(1.35)
3.86
(1.11)
6.01
(2.08)
3.55
(1.00)
3.12
(1.03)
5.59
(1.91)
6.22
(1.78)
7.29
(2.28)
15.8
(4.90)
10.3
(3.12)
6.95
(2.05)
13.1
(3.83)
6.21
(1.91)
22.7
(9.34)
11.6
(3.74)
8.17
(2.06)
20.3
(6.83)
4.76
(1.34)
6.73
(2.39)
3.68
(1.06)
9.52
(3.00)
4.50
(1.36)
6.37
(1.92)
23.2
(8.59)
5.92
(1.72)
5.66
0.5
>220
-
2
170
12.4
-
-
-
(147-192)
108
(3.80)
3.69
5
1
1
>54
-
1
8.40
6.39
8.61
11.1
6.55
5.58
2.33
11.0
4.76
13.9
8.04
10.3
3.32
22.0
9.83
11.9
8.09
5.22
6.45
8.37
<9.02
8.37
7.82
10.3
4.47
11.6
7.11
3.28
5.03
11.3
20.8
6.52
4.85
1.68
-
-
>350
-
-
(71.5-158)
293
(0.96)
5.78
6
1
1
-
2
-
-
(210-379)
96.8
(1.77)
11.5
7
0.25
2
0.25
2
57.9
1.70
(0.54)
-
1
1.67
-
(30.1-94.0)
>200
-
(77.6-114)
125
(4.08)
3.92
9
2
-
(81.1-175)
94.8
(1.10)
3.17
10
11
13
18
19
23
25
26
30
32
34
35
49
51
52
53
54
55
56
57
71
73
76
79
80
81
82
1
0.5
0.5
0.5
2
72.8
4.27
(1.34)
-
2
1.30
-
(49.0-99.1)
>100
-
(59.9-156)
68.8
(1.09)
35.0
0.5
0.5
2
1
-
(64.5-72.7) (11.9)
41.2
3.53
(0.91)
-
2
64.5
(42.0-84.7)
36.3
4.54
(1.36)
13.3
1.57
-
(27.0-56.7)
>45
1
-
-
(32.1-40.0)
62.5
(4.31)
15.5
1
1
>68
-
1
-
-
-
(55.6-67.9)
204
(4.84)
4.29
1
1
88.6
1.87
(0.57)
2.80
(1.02)
-
2
2.30
1.10
-
(45.5-174)
33.6
(21.4-57.7)
(132-271)
37.0
(1.31)
11.9
0.5
1
0.25
1
(30.7-45.7)
356
(4.76)
4.04
0.25 >350
2
-
>100
-
79.6
(44.2-139)
86.9
(71.5-109)
>150
-
43.3
(18.4-112)
>80
-
12.8
(8.25-18.6)
34.4
(24.2-48.2)
>70
-
-
(242-547)
34.0
(1.52)
2.75
2
2
-
2
-
-
(20.7-51.6)
243
(0.75)
6.02
1
1
2.16
(0.65)
11.4
(4.49)
-
4
3.05
1.51
-
(172-305)
131
(2.04)
6.47
4
1
4
(111-159)
106
(1.70)
6.31
4
4
4
-
(85.1-143)
73.5
(1.98)
10.5
0.25
1
0.25
1.54
(0.57)
-
1
1.70
-
(64.3-86.4)
60.7
(3.08)
45.2
0.25
2
-
(58.9-63.8) (14.5)
1
2
3.34
(1.16)
4.38
(1.45)
-
2
35.3
(25.0-43.4)
76.5
6.78
(2.05)
10.5
(3.10)
-
2.75
2.23
-
2
2
2
(68.9-92.5)
4
4
2
<100b
-
2
2
>110
-
-
2
104
26.9
-
-
(100-113)
52.5
(8.99)
8.47
4
4
>75
-
4
-
-
-
(45.0-63.7)
57.9
(2.46)
6.43
1
1
18.7
4.26
(1.47)
-
1
3.09
-
(12.6-31.1)
>100
(43.9-75.3)
113
(2.23)
17.4
0.5
1
0.5
1
1
(103-123)
181
(5.73)
4.59
>46
-
2
-
-
-
(123-251)
88.0
(1.27)
24.0
1
1
>120
-
2
-
-
-
(83.3-94.9)
53.2
(6.85)
5.90
1
1
>120
-
2
-
-
-
(41.4-72.5)
123
(1.89)
6.92
2
2
>200
-
2
-
-
41.9
-
(102-150)
43.2
(2.10)
6.26
1
0.5
1
7.72
(2.56)
-
1
1.03
-
(1.73)
3.68
(0.99)
11.2
(3.52)
20.8
(7.15)
7.31
(34.6-50.9)
>300
-
>50
-
>50
-
209
(32.5-58.9)
197
(1.80)
12.8
1
4
(6.35-13.0)
6.32
(5.44-7.23)
9.85
(8.77-10.7)
287
(237-359)
-
(174-227)
41.2
(3.96)
14.4
phenytoin
1
1
-
0.5
0.25
0.25
-
-
(36.9-46.1)
47.8
(4.82)
7.98
carbamazepine 0.25
0.25
0.25
-
-
-
(39.2-59.2)
483
(2.37)
12.3
valproate
0.25
8.51
(2.69)
2.31
(2.48)
(176-249)
(412-571)
(4.01)
a
b
PI indicates the protection index, i.e. TD50/ED50
mg/kg, respectively.
.
Toxicity was displayed in 0/8, 3/8, and 8/8 mice using doses of 50, 70, and 100

(Aryloxy)aryl Semicarbazones and Related Compounds
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 20 3991
Ta ble 3. Evaluation of Selected Compounds in the MES and Neurotoxicity Tests after Oral Administration to Rats
MES screen
95% CI
neurotoxicity screen
compd
t
ED50
slope
SE
t
TD50
95% CI
slope
SE
PIa (MES)
5
6
8
9
2
4
2
4
4
4
2
1
2
6
2
4
4
4
4
0.5
2
2
4
2
4
4
2
4
1.59
6.15
11.4
5.53
2.37
1.13
5.65
3.07
3.43
6.48
2.63
3.21
1.68
45.8
9.73
23.1
3.37
4.25
2.92
2.89
1.52
4.39
43.4
4.29
16.6
18.6
4.98
9.11
3.65
18.7
23.2
3.57
395
1.01-2.25
3.69-9.17
7.61-15.8
3.18-8.75
1.54-3.62
0.71-2.01
3.79-7.81
2.58-3.94
2.28-4.73
2.97-15.5
1.69-3.93
2.25-4.64
1.15-2.44
19.5-316
6.44-14.1
14.3-36.6
2.37-4.72
2.89-5.97
2.20-3.46
1.57-5.29
0.99-2.30
2.67-5.83
25.1-66.3
3.20-5.24
13.7-18.9
14.2-25.0
3.24-7.01
6.19-11.7
2.42-5.30
12.4-27.6
21.4-25.4
2.41-4.72
332-441
3.17
2.55
4.12
2.44
3.18
2.66
3.65
7.11
4.12
1.98
3.21
3.58
4.44
1.33
3.84
3.14
5.74
3.67
5.77
2.04
3.60
4.21
2.29
6.02
11.8
5.24
3.92
5.29
3.84
3.93
15.1
3.84
8.13
0.84
0.69
1.32
0.66
0.81
0.95
0.98
2.29
1.32
0.75
0.82
1.02
1.28
0.52
1.30
0.92
1.80
1.04
1.60
0.59
1.02
1.28
0.57
2.00
4.49
1.67
1.10
1.50
1.30
1.11
4.28
1.15
2.76
0.25-24b
>500
-
-
-
-
-
>315
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
-
0.25-24
>500
>500
>90
>500
>500
>500
-
-
-
>90
>210
>79
>88
>163
>146
-
10
18
21
22
23
30
32
34
35
45
49
50
51
52
53
54
55
56
61
73
74
75
76
77
82
90
0.25-24
-
-
2
-
-
0.25-24
-
-
0.25-24
-
-
0.25-24
-
-
-
-
-
0.25-24
>500
>322
>500
>100
-
-
-
>190
>100
>297
>2.18
-
0.25-24
-
-
0.25-24
-
-
c
-
-
-
-
-
-
-
109
-
-
-
2
80.3-178
4.82
-
1.82
32.3
>16.9
-
4
4
>72
<500d
>500
>500
-
-
>496
-
-
183
-
>500
>125
>500
361
-
-
-
-
-
-
-
-
-
-
-
-
0.25-24
-
-
>173
>328
-
-
>116
-
0.25-24
-
-
-
-
-
-
-
-
0.25-24
-
-
1
2
2
2
2
0.5
2
1
0.5
-
-
-
-
-
-
-
36.8
-
4
101-338
2.49
-
0.86
-
-
-
-
-
-
2.96
2.17
0.25-24
-
-
>137
>6.70
>21.6
101
2.17
2
-
-
phenytoin
carbamazepine
valproate
0.25-24
1
0.5
-
-
319-402
719-1148
11.4
6.57
859
a
b
PI indicates the protection index, i.e. TD50/ED50
.
The compound was examined 0.25, 0.5, 1, 2, 4, 6, 8, and 24 h after administration
except for 22, in which case observation at the end of 8 h was omitted. c No toxicity was noted at the end of 0.25, 0.5, 1, 2, and 4 h
after administration. Toxicity was observed in 0, 0, 0, 3, 5, 2, 1, and 0 rats out of 8 animals 0.25, 0.5, 1, 2, 4, 6, 8, and 24 h after
d
administration.
A positive correlation between the MES activity of the
compounds in point v, i.e. 5, 51, and 52 was noted with
both the MR and SA constants, i.e. activity increased
as the size of these physicochemical constants rose. One
may conclude therefore that the hypothesis whereby a
hydrophobic area on the binding site exists with which
the alkyl groups of 51 and 52 interact is strengthened.
In the case of point vi, the r values for the linear plots
of the MES ED50 figures against the MR and SA
constants were 0.693 and 0.709, which shows a general
trend whereby activity increased as the MR and SA
constants were elevated. In the remaining cases, no
correlations were noted.
With few exceptions, the compounds were adminis-
tered orally to rats and examined in the MES screen.
These data are portrayed in Table 1. Initially a dose of
50 mg/kg was employed. However under these circum-
stances, virtually all of the compounds afforded complete
protection, and hence the dose was reduced principally
4-fold in order to detect candidate anticonvulsants with
marked potencies. The data in Table 1 reveal that in
most cases complete protection was displayed by the
rats which received 12.5 mg/kg of the compounds.
Using principally the doses listed in Table 1, 18 com-
pounds were examined in the scPTZ screen of which half
were inactive, and in general the remaining analogs had
marginal potencies whereby 25% of the rats were
protected against seizures. With the exception of 19,
which caused neurological deficit in one of four rats,
using the doses listed in Table 1, all of the compounds
F igu r e 3. ORTEP diagram of 1.
in Table 1 which were examined in the MES screen were
bereft of neurotoxicity.
In addition, this qualitative examination revealed
that complete protection was demonstrated in either
all or the majority of the compounds in the series
2-48, 49-57, 73-83, and 93, 94, suggesting that
these molecules are accommodated well at the bind-
ing site. On the other hand, the diminished activity
of the compounds in the series 58-66 reinforces the
belief that alignment at the binding site is reduced
when the aryloxy group is placed in the ortho position

3992 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 20
Dimmock et al.
F igu r e 7. ORTEP diagram of 73.
49-57, and 73-83. The data are summarized in Table
3. Evaluation in the scPTZ screen was undertaken with
63% of the semicarbazones and related compounds
which are listed in Table 3. At the maximum doses ad-
ministered, the majority of the compounds were inactive
while a few demonstrated very marginal anticonvulsant
properties. The only exception was 21, which afforded
protection in 50% of the rats using a dose of 50 mg/kg.
The marked MES selectivity displayed by these com-
pounds is thus reinforced, and further discussion will
refer to their evaluation in the MES screen only. Neuro-
toxicity was absent in virtually all compounds at the
highest doses employed. The data in Table 3 revealed
the very high potencies of many of these compounds in
the rat oral MES screen. In fact the desired goal of an
ED50 figure of 2-3 mg/kg or less was achieved with 5,
10, 18, 32, 35, and 53-55. The potential importance
of these compounds as anticonvulsants is further em-
phasized by the following two considerations. First,
bearing in mind that the preferred route of administra-
tion of drugs is oral, especially for long-term therapy,
the extremely high PI values displayed by most of these
compounds suggests a wide margin of safety. Second,
a comparison with three established drugs was made.
The data reveal that 93% of the compounds listed in
Table 3 had ED50 figures lower than and 84% had PI
values greater than those of phenytoin; the analogous
figures for carbamazepine were 40 and 53%, respec-
tively. All of the compounds were more potent than
valproic acid and had higher PI values.
F igu r e 4. ORTEP diagram of 2.
F igu r e 5. ORTEP diagram of 5.
The marked increases in potencies of 5 and 51
compared to 90 and 61, respectively, indicated the
importance of both the primary amino group and also
the position of the aryloxy function in the proximal aryl
ring in conferring anticonvulsant activity. One of the
earliest compounds to be evaluated in the rat oral MES
screen was 5, and currently it is undergoing extensive
bioevaluation, details of which will be presented else-
where.
Linear and semilogarithmic plots were made between
the ED50 figures in the MES screen and the σ,σ*, π, MR,
and SA figures of (i) all of the compounds in series 2-48
listed in Table 3; (ii) 49, 51, 53, and 55; (iii) 50, 52, 54,
and 56; and (iv) 73, 76, and 82. In addition, an
evaluation of the importance of the size and shape of
the groups attached to the azomethine carbon atom and
F igu r e 6. ORTEP diagram of 23.
of the distal aryl ring. Other structural alterations
whereby the size of the spacer group was greater than
oxygen and sulfur atoms 67-72 or replacements of
the oxygen and primary amino groups of the semicar-
bazono functions 84-87, 88-92 appeared to be detri-
mental.
Quantitation in the rat oral MES screen was under-
taken with approximately one-third of the compounds
described in this study, principally from the series 2-48,

(Aryloxy)aryl Semicarbazones and Related Compounds
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 20 3993
the para position of the distal aryl ring were found by
plotting the MES ED50 figures against the MR and SA
constants found in (v) 5, 51, and 52 and (vi) 5, 23, 30,
32, 34, 35, and 45.
The compounds in cases ii and iv correlated positively
with the σ, π, MR, and SA constants, except a negative
correlation was noted between the σ values and MES
ED50 figures in iv. The potencies of the semicarbazones
listed in iii correlated positively with the σ and SA
constants. A plot of the MR and SA constants of 5, 51,
and 52 (v) against the MES ED50 figures in Table 3
somewhat surprisingly revealed a negative correlation.
No correlations were observed in the remaining cases.
Thus relationships were noted with various compounds
in the series 49-57 and 73-83 but not 2-48. In
general, increases in the σ, π, MR, and SA constants in
49-57 and 73-83 caused an increase in activity, and
this observation can be utilized in subsequent drug
design. However the exception to this general trend is
the group of compounds 5, 51 and 52 (v) whereby
activity was diminished with increasing size and shape
of the substituents. Hence future work should be
directed to placing various groups on the azomethine
carbon atom in order to evaluate further the structural
requirements for interaction at the binding site.
F igu r e 8. Orientations of the proximal and distal aryl rings
of 1 (colorless), 2 (red), 5 (green), 23 (orange), 73A (blue), and
73B (yellow) when the N3, C14, (O2 or O), and N2 atoms of
each molecule are superimposed.
and 3 reveal that the times of peak effect of 5, 51, and
52 in the mouse and rat MES screens were 1 and 2 h,
Since bioactivity is considered to be influenced by the
rate and extent of the passage of a drug to its site of
action,5 the partition coefficients between 1-octanol and
buffer, pH 7.4, of 10 representative compounds and
carbamazepine were determined. The log P values of
these compounds are indicated in parentheses, viz. 1
(2.07), 2 (2.66), 5 (1.91), 10 (2.76), 44 (1.29), 47 (1.92),
51 (2.03), 76 (2.80), 90 (1.96), 91 (2.06), and carbam-
azepine (2.20). The data support the concept that the
compounds align at a specific binding site and are not
structurally nonspecific.6 First, 1, which is either very
weakly active or inactive in the mouse intraperitoneal
and rat oral MES screens (Table 1) has a partition
coefficient similar to that of 5, 51, 90, or 91, which
demonstrated marked anticonvulsant activity. Second,
linear and semilogarithmic plots between the partition
coefficients and mouse intraperitoneal ED50 figures of
5, 10, 51, and 76 did not show any correlation (r < 0.8).
Similarly no relationship was established between the
partition coefficients of 5, 10, 51, 76, and 90 and the
ED50 rat oral data. Third, the mouse and rat data for
90 and 91 in Table 1 indicated the greater anticonvul-
sant activity of 90, yet both have similar partition
coefficients. Fourth, in the MES screens, the isosteres
5 and 76 had the same activity in the mouse intraperi-
toneal test and both compounds are highly potent when
given orally to rats, yet their partition coefficients are
markedly divergent. Finally it is of interest to note that
the average log P value for all of the active compounds
was 2.15, which is similar to the figure for carbam-
azepine.
1
respectively. Hence H NMR spectra were recorded at
dissolution and 3 h after incubation of the solutions at
37 °C. No changes in the spectra were noted. Since
the stereochemistry of the carbimino group in different
semicarbazones was shown by X-ray crystallography to
have the E configuration (vide infra), the compounds
were considered to retain this stereochemistry in vivo.
The second phase of the study was an evaluation of
the binding site hypothesis using X-ray crystallography.
Examination of the shapes of molecules under nonbio-
logical conditions does not necessarily reflect in vivo
situations. In the case of X-ray crystallography, the
structures were solved when the compounds were solids,
and their flexibilities may permit various orientations
to occur. Nevertheless, the data obtained may afford
insight into their likely shapes in vivo. Results were
obtained for four compounds which displayed activity
in the MES mouse intraperitoneal and rat oral screens
(2, 5, 23, 73) and one semicarbazone which was very
weakly active or inactive at the doses employed in both
tests (1). The ORTEP diagrams of 1, 2, 5, 23, and 73
are portrayed in Figures 3-7, respectively. In the case
of 73, the C5, C4, S, C8, C9, C11, and C12 atoms occupy
two different positions, each with an occupancy of 0.5.
These two molecules are referred to as 73A and 73B.
The carbimino group had the E configuration in all five
compounds.
In order to compare the shapes of the five compounds,
four atoms in the semicarbazono group, namely the N3-
C14-(O2 or O)-N2 atoms, were superimposed (Figure
8). The results show that in the crystal state while the
proximal ring occupies a similar position in all five
compounds, the distal aryl rings are found in three
distinct locations which may be designated as location
A (2 and 73), location B (5 and 23), and location C (1).
The determination of the precise positions of the aryl
rings from the X-ray crystallographic data of all five
compounds was undertaken in order to obtain further
information pertaining to the nature of the binding site.
The distances between the C14 atom and the centers of
In solution, semicarbazones are capable of displaying
E/Z isomerization pertaining to the carbimino double
bond. The ratio of isomers could be influenced by the
sizes of the groups on the carbimino carbon atom,7 and
hence the three representative compounds chosen,
namely 5, 51, and 52, reflected this possibility. Due to
the very low aqueous solubilities of the compounds
prepared in this study, 10 mM solutions were prepared
in deuterated dimethyl sulfoxide. The data in Tables 2

3994 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 20
Dimmock et al.
(O2 or O)-N2 plane. Thus with the exception of 2, there
was a greater displacement of the distal than the
proximal rings from the ureido group in 1, 5, 23, and
73. The distal aryl ring in 5, which had the lowest ED50
figures in both the quantitative MES screens, was
located above the N3-C14-(O2)-N2 plane while the
marginally less active compounds 23 and 73 had distal
rings below this plane. This observation may indicate
that there is a pocket at location B and the distal ring
of 5 could interact with the top of this cavity. The X-ray
data have therefore provided further insight into the
putative binding site of these novel anticonvulsants.
a
b
Con clu sion s
This study has revealed that many (aryloxy)aryl
semicarbazones and related compounds have marked
potency in the rat oral MES screen and did not dem-
onstrate neurotoxicity at the highest doses adminis-
tered. From the QSAR data, the size and shape of
various groups in these molecules correlated positively
with activity in the MES screen while the use of π
constants, as well as partition coefficient determina-
tions, indicate that in general the hydrophobicity of the
molecules did not influence anticonvulsant activity. The
X-ray crystallographic data suggested that the distal
aryl ring occupied different positions at the distal
binding site and that certain interatomic distances and
bond angles affected potency. The deductions from the
screening and X-ray crystallographic results suggest
that the putative binding site at which MES-active
compounds interact is a valid hypothesis which will be
scrutinized carefully as additional data is generated.
F igu r e 9. (a) Distances and θ angles between the C14 atom
and centers of both the proximal (CArp) and distal (CArd) rings
in 2. (b) Displacement and ψ angles of the centers of the
proximal (CArp) and distal (CArd) rings from the N3-C14-
(O2)-N2 plane in 2.
both aryl rings were measured. In addition, the angles
θ between the centers of the aromatic rings and the
C14-N2 bonds were obtained. Furthermore the dis-
placements of the aryl rings above or below the N3-
C14-(O2 or O)-N2 plane were calculated for each
compound in terms of both distances and angles ψ. The
determination for a representative compound 2 is il-
lustrated in Figure 9.
The results revealed that the distances between the
C14 atom and the centers of the proximal aryl rings (dp)
are similar for all compounds, namely in the range 6.0-
6.1 Å. On the other hand, the dd figures for the
compounds in locations A-C are 10.9-11.1, 10.1-10.2,
and 8.5 Å, respectively. If these spatial arrangements
are the ones which align at the binding site, the fact
that the MES figures in both the mouse ip and rat oral
screens of 5 and 23 are lower than 73 may indicate that
the dd distances affect anticonvulsant activities. Thus
the preparation of rigid analogs of the (aryloxy)aryl
semicarbazones in which the distal aryl ring is confined
to location B may lead to compounds with increased
potencies.
The θp angles were between 29° and 34°. The
synthesis and anticonvulsant evaluation of analogs in
which this angle is altered, such as by replacing the
hydrogen atom of the carbimino group by different
functions, may reveal the importance of this angle in
conferring anticonvulsant activity. The θd angles of 1,
2, 5, 23, 73A, and 73B were 64°, 25°, 51°, 51°, 28°, and
28°, respectively, and hence the θd figures for 5 and 23
in location B are approximately twice those for 2 and
73 in location A. However a further increase in the θd
angle from approximately 51° to that found in 1 is
detrimental to activity.
Exp er im en ta l Section
A. Ch em istr y. Melting points are uncorrected. 1H NMR
spectra were recorded routinely using a Varian T-60 spec-
trometer (60 MHz) while the stability studies used a Bruker
AM 300FT instrument (300 MHz). Partition coefficients were
determined using a Gilford UV/vis spectrophotometer. Thin
layer chromatography (TLC) was performed using silica gel
sheets with a fluorescent indicator. Elemental analyses (C,
H, N) were undertaken by Mr. K. Thoms, Department of
Chemistry, University of Saskatchewan, and in most cases
were within 0.4% of the calculated values. For the following
compounds, despite having sharp melting points and display-
ing homogeneity by TLC, repeated recrystallizations and
submission for elemental analyses revealed that the figure for
one element deviated slightly from (0.4%, namely (element,
difference between the calculated and found values expressed
as a percentage) 4 (H, 0.48), 8 (N, 0.46), 11 (C, 0.63), 21 (N,
0.65), 37 (C, 0.84), 58 (N, 0.47), 59 (C, 0.55), 71 (N, 0.53), 81
(C, 0.66), and 95 (N, 0.55). In all cases except 11 and 81, the
values determined were lower than the calculated figures,
which may have been due to a problem of incomplete combus-
tion.
Syn th esis of Com p ou n d s in th e Ser ies 1, 2-48, 49-57,
58-66, 73-83, 93, 94, a n d 95. 3-Phenoxybenzaldehyde
required in the synthesis of 1 was obtained from the Aldrich
Chemical Co., Milwaukee, WI. The intermediate (aryloxy)-
aryl and (arylthio)aryl aldehydes required in the synthesis of
the other compounds were prepared as follows.
The data obtained from the measurements indicated
in Figure 9b indicated the displacement of the proximal
and distal aryl rings from the N3-C14-(O2 or O)-N2
plane. The proximal rings are nearly coplanar with the
ureido group as revealed by the d′p and ψp figures. The
d′d figures for 1, 2, 5, 23, 73A, and 73B were 2.3, 0.02,
0.71, -0.82, -1.4, and -1.4 Å, respectively, the negative
figures indicating that the ring was below the N3-Cl4-
Anhydrous potassium carbonate (0.12 M) was added to a
solution of the appropriate phenol or thiophenol (0.15 M) and
4-fluorobenzaldehyde, 4-fluoroacetophenone, or 4-fluoropro-
piophenone (0.14 M) in dimethylacetamide (100 mL). For 95,
4-fluorobenzaldehyde (0.28 M) was taken. The mixture was
heated under reflux at 155 °C under nitrogen, and the progress
of the reaction was monitored by TLC using a solvent system
of benzene:methanol (9:1). After ∼5-10 h, the mixture was
cooled and water (100 mL) was added. The reaction mixture

(Aryloxy)aryl Semicarbazones and Related Compounds
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 20 3995
was extracted with chloroform (2 × 100 mL), and the combined
organic extracts were washed with aqueous sodium hydroxide
solution (4% w/v) and water. After the mixture was dried over
anhydrous magnesium sulfate, the solvent was removed in
vacuo and the resultant oil was distilled under reduced
pressure to give the appropriate (aryloxy)aryl or (arythio)aryl
aldehyde or ketone. The purity of the distillate was checked
solutions were made using 1-octanol to which buffer was
added. The λmax and ꢀ values of the compounds were obtained
in 1-octanol and not phosphate-buffered saline, pH 7.4, due
to the low aqueous solubilities of the compounds. The log P
values (error in parentheses) were as follows: 1, 2.07 (0.08);
2, 2.66 (0.11); 5, 1.91 (0.08); 10, 2.76 (0.11); 44, 1.29 (0.05);
47, 1.92 (0.08); 51, 2.03 (0.08); 76, 2.80 (0.11); 90, 1.96 (0.08);
91, 2.06 (0.08); and carbamazepine, 2.20 (0.09).
1
by TLC using benzene:methanol (9:1) as the solvent. The H
NMR spectral data (300 MHz) of a representative intermedi-
ate, namely 4-phenoxybenzaldehyde, were as follows: δ (CDCl3)
9.94 (s, 1H, CHO), 7.82-7.88 (2t, 2H, ortho H of proximal aryl
ring), 7.38-7.46 (m, 2H, meta H of proximal aryl ring), 7.20-
7.27 (m, 1H, para H of distal aryl ring), 7.03-7.12 (m, 4H,
ortho and meta H of distal aryl ring).
Sta tistica l An a lyses. The σ, π, and MR constants were
taken from the literature,14 and the Taft σ* values were
obtained from a reference source.15 The solvent accessible
surface areas calculated with a probe radius of 1.4 Å were
obtained using a MacroModel Version 4.5 program16,17 and a
Silicon Graphics Indigo Extreme workstation. The MES
figures recorded in Tables 2 and 3 which were used in the
statistical analyses were converted to µmol/kg. The correla-
tions observed were summarized as follows: group of com-
pounds (refer to discussion), physical constant, and linear (rl)
and semilogarithmic (rsl) correlation coefficients. For the
mouse intraperitoneal screen (Table 2) the following correla-
tions were established: (v) MR, rl ) 0.955, rsl ) 0.931; (v) SA,
r1 ) 0.919, rsl ) 0.899. In the case of the rat oral test, the
data summarized refer to the group of compounds, physical
constant, rl and rsl figures: (ii) σ, 0.808, 0.825; (ii) π, 0.786,
0.851; (ii) MR, 0.692, 0.819; (ii) SA, 0.899, 0.926; (iii) σ, 0.765,
0.817; (iii) SA, 0.834, 0.861; (iv) σ, 0.998, 1.000; (iv) π, 0.847,
0.865; (iv) MR, 0.956, 0.966; (iv) SA, 0.778, 0.799; (v) MR,
0.971, 0.947; and (v) SA, 0.990, 0.975.
X-r a y Cr ysta llogr a p h y of 1, 2, 5, 23, a n d 73. The
compounds were crystallized from ethanol:dimethyl sulfoxide
(1), ethyl acetate:cyclohexane (2), acetone:methanol (23), and
hexane:methanol (73) by vapor diffusion while 5 was crystal-
lized from hot ethanol. An Enraf-Nonius CAD-4 diffractometer
with an ω scan was used for data collection, and the structure
was solved by direct methods using NRCVAX.18 Atomic
scattering factors were taken from the literature.19 All non-
hydrogen atoms were found on the E-map and refined aniso-
tropically. Hydrogen atom positions were calculated and not
refined. In the case of 73, the atoms S, C3, C4, C8, C9, C11,
and C12 occupied two different positions, each with an
occupancy of 0.5.
A mixture of semicarbazide hydrochloride (0.01 M), sodium
acetate (0.01 M), and water (10 mL) was added slowly to a
stirring solution of the (aryloxy)- or (arylthio)aryl aldehyde
(0.01 M) in ethanol (95%, 30 mL). The reaction mixture was
stirred at room temperature for 1-2 h, and the precipitate
was collected, washed with ether, and dried. Most of the
compounds were recrystallized from ethanol (95% v/v), except
for compounds 2, 3, 5, 9-16, 44, 45, 47, and 48 (absolute
1
ethanol) and compound 23 (methanol). The H NMR spectral
data (300 MHz) of a representative compound 5 in dimethyl-
d6 sulfoxide were as follows: 9.40 (s, 1H, NH), 6.94 (s, 1H,
CH), 6.82-6.88 (2t, 2H, meta H of distal aryl ring), 6.33-6.42
(m, 2H, ortho H of distal aryl ring), 6.05-6.30 (m, 4H, ortho
and meta H of proximal aryl ring), 5.60 (s, 2H, NH2). The
literature melting points (°C) of 1, 2, and 58 were 217,8 219-
220,9 and 219-220,10 respectively. The percentage yields are
based on the last reaction between the (aryloxy)aryl or
(arylthio)aryl carbonyl compounds and semicarbazide.
Syn th esis of Com p ou n d s in th e Ser ies 67-72. 4-(Ben-
zyloxy)benzaldehyde required in the synthesis of 67 was
obtained from the Aldrich Chemical Co., Milwaukee, WI. The
other intermediate aldehydes were prepared as follows.
Benzoyl chloride or 4-chlorobenzoyl chloride (0.05 M) was
added to a solution of 4-hydroxybenzaldehyde (0.04 M) in
pyridine (100 mL). After standing overnight at room temper-
ature, the reaction mixture was poured onto acetic acid (2 N,
100 mL). The precipitate was collected, washed with water,
and recrystallized from water-methanol to give 4-(benz-
oyloxy)benzaldehyde and 4-[(4-chlorobenzoyl)oxy]benzalde-
hyde required in the synthesis of 68 and 69, respectively.
4-(Phenylsulfonyl)benzaldehyde used in the synthesis of 70
was prepared as follows. A mixture of sodium benzenesulfi-
nate (0.11 M) and 4-fluorobenzaldehyde (0.1 M) in dry di-
methyl sulfoxide (75 mL) was stirred at 100 °C for 18 h under
nitrogen and then poured onto ice (∼200 g). The precipitate
was collected, washed with water, and recrystallized from
ethanol (95% v/v). Finally, benzenesulfonyl chloride or 4-
methylbenzenesulfonyl chloride (0.20 M) was added dropwise
to a stirred solution of 4-hydroxybenzaldehyde (0.16 M) in di-
chloromethane (90 mL) and trimethylamine (3-5 mL) at 0-10
°C over a period of 10 min. After a further 15 min, the reaction
mixture was diluted with dichloromethane and successively
extracted with water, hydrochloric acid (10% w/v), saturated
sodium bicarbonate solution, and saturated sodium chloride
solution. After the organic extract was dried, the solvent was
removed, affording the products required in the syntheses of
71 and 72. The compounds were homogeneous by TLC using
a solvent system of benzene:methanol (7:3), and the melting
points of 68, 70 - 72 were in accord with literature values.
These intermediate aldehydes were reacted with semicar-
bazide as described previously.
The data for 1 were as follows: C14H13N3O2, Mr ) 255.27, a
) 11.809(3) Å, b ) 7.1862(20) Å, c ) 15.4816(20) Å, â )
109.220(20), Z ) 4, space group ) P21/a, monoclinic, Dx ) 1.367
g cm-3, λ(Mo KR) ) 0.7093 Å, T ) 123 K. Merging R is based
on intensities 0.012 for 480 replicate reflections, R(F) ) 0.046,
Rw ) 0.049, S ) 3.82. A total of 2652 reflections were
measured, 2172 of which were independent. The refinement
of the structure used 1728 observed reflections [I > 2σ(I)].
Parameters refined ) 172 [w ) 1/σ2(F)]. ∆F in the final
difference map was within +0.25 and -0.240 e Å-3
.
The data for 2 were as follows: C14H13N3O2, Mr ) 255.27, a
) 12.4637(15) Å, b ) 7.6840(22) Å, c ) 12.9457(15) Å, â )
91.321(10), Z ) 4, space group ) P21/c, monoclinic, Dx ) 1.368
g cm-3, λ(Mo KR) ) 0.7093 Å, T ) 287 K. Merging R is based
on intensities 0.007 for 316 replicate reflections, R(F) ) 0.039,
Rw ) 0.044, S ) 2.33. A total of 2489 reflections were
measured, 2173 of which were independent. The refinement
of the structure used 1670 observed reflections [I > 2σ(I)].
Parameters refined ) 172 [w ) 1/σ2(F) + 0.0001]. ∆F in the
final difference map was within +0.24 and -0.260 e Å-3
.
The data for 5 were as follows: C14H12N3O2F, Mr ) 273.27,
a ) 12.7066(19) Å, b ) 7.7389(15) Å, c ) 13.2838(23) Å, â )
103.965(15), Z ) 4, space group ) P21/a, monoclinic, Dx ) 1.432
g cm-3, λ(Mo KR) ) 0.7093 Å, T ) 123 K. Merging R is based
on intensities 0.017 for 301 replicate reflections, R(F) ) 0.038,
Rw ) 0.051, S ) 2.15. A total of 2508 reflections were
measured of which 2207 were independent. The refinement
of the structure used 1750 observed reflections [I>2σ(I)].
Parameters refined ) 181 [w ) 1/σ2(F) + 0.0002]. ∆F in the
Syn th esis of Com p ou n d s in th e Ser ies 84-87 a n d 88-
92. These compounds were prepared from the appropriate
(aryloxy)aryl and (arylthio)aryl aldehydes using literature
methodologies.11,12 The time of heating the reactants under
reflux was 6 h (84-87), while the times of stirring the
reactants at room temperature were 8 (88), 10 (89, 91), and
14 (92) h. In the case of 90, the reaction mixture was heated
at 60 °C for 0.5 h. All compounds were recrystallized from
ethanol (95% v/v).
final difference map was within +0.17 and -0.260 e Å-3
.
The data for 23 were as follows: C15H15N3O2, Mr ) 269.30,
a ) 13.0692(6) Å, b ) 7.8606(4) Å, c ) 13.5724(6) Å, â )
101.857(4)°, Z ) 4, space group ) P21/a, monoclinic, Dx ) 1.311
g cm-3, λ(Mo KR) ) 0.7093 Å, T ) 287 K. Merging R is based
on intensities 0.006 for 119 replicate reflections, R(F) ) 0.041,
Deter m in a tion of log P va lu es. The log P figures were
determined by a previously reported procedure13 except that

3996 J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 20
Dimmock et al.
Rw ) 0.056, S ) 2.51. A total of 2513 reflections were
measured, 2394 of which were independent. The refinement
of the structure used 1858 observed reflections [I > 2.5σ(I)].
Parameters refined ) 181 [w ) 1/σ2(F) + 0.0002]. ∆F in the
time of activity), viz. 12 (1, 4 h), 17 (1, 0.5, 1, 4 h), 21 (1, 0.5,
1, 2 h), 25 (1, 0.25, 4 h), 27 (1, 0.25, 1, 4 h, and 4, 2 h), 34 (1,
0.25, 1, 4 h, and 2, 2 h), 52 (1, 0.5, 1 h), 67 (1, 1 h), and 87 (1,
4 h). Using the doses indicated in Table 1, no neurological
deficit was noted except for 19, whereby one of four rats
demonstrated neurotoxicity 1, 2, and 4 h after administration.
In the case of 27 and 42, the dose in both the scPTZ and NT
screens was 50 mg/kg.
Quantitative evaluation of selected compounds in the rat
oral MES and NT screens was undertaken, and the data are
presented in Table 3. The following compounds were exam-
ined in the rat oral scPTZ test, and the results are as follows
final difference map was within +0.22 and -0.25 e Å-3
.
The data for 73 were as follows: C14H13N3OS, Mr ) 271.32,
a ) 12.9365(15) Å, b ) 5.3111(11) Å, c ) 19.061(4) Å, â )
98.46(3), Z ) 4, space group ) P21/a, monoclinic, Dx ) 1.391
g cm-3, λ(Mo KR) ) 0.7093 Å, T ) 123 K. Merging R is based
on intensities 0.012 for 118 replicate reflections, R(F) ) 0.062,
Rw ) 0.078, S ) 2.95. A total of 2394 reflections were
measured, 2276 of which were independent. The refinement
of the structure used 1519 observed reflections [I > 2.5σ(I)].
Parameters refined ) 235 [w ) 1/σ2(F)]. ∆F in the final
at the maximum dose in mg/kg, time of administration
(ED50
in hours): 5 (>250, 2), 9 (>250, 4), 10 (>250, 4), 18 (>250,4),
22 (>250, 1), 23 (>125, 2), 32 (>250, 2), 34 (>161, 4), 35 (>250,
4), 51 (>250, 2), 52 (>250, 2), 53 (>250, 4), 54 (>250, 2), 55
(>250, 4), 73 (>248, 4), 76 (>250, 2), 82 (>250, 2), and 90
(>250, 0.5). In the case of 21, the number of rats protected
(out of four) at the end of 0.25, 0.5, 1, 2, and 4 h were 0, 0, 0,
1, and 2. The semicarbazones 6, 8, 30, 45, 49, 50, 56, 61, 74,
75, and 77 were not evaluated in the rat oral scPTZ and NT
screens. The figures for three reference drugs in the rat oral
scPTZ screen were as follows: phenytoin (>250, 2 h), carbam-
azepine (>250, 1 h), and valproic acid (620, 0.5 h). For the
latter compound, the 95% CI and slope were 469-985 mg/kg
and 3.17, respectively.
difference map was within +0.31 and -0.31 e Å-3
.
B. P h a r m a cology. The anticonvulsant evaluations were
undertaken by the National Institute of Neurological Disorders
and Stroke, National Institutes of Health, using their reported
procedures.20
(i) In tr a p er iton ea l In jection in Mice. All compounds
were examined initially in the MES, scPTZ, and NT screens,
and these data are presented in Table 1. Side effects were
noted in the scPTZ screen for the following compounds at
various doses (mg/kg) and time intervals. Continuous seizure
activity (CSA) was noted in the case of 11 (300, 0.5 h, 4 h), 13
(300, 0.5 h), 14 (300, 0.5 h, and 100, 300, 4 h), 16 (100, 300,
0.5 h, 4 h), 18 (300, 0.5 h, and 100, 300, 4 h), 19 (100, 300, 0.5
h, and 300, 4 h), 22 (100, 300, 0.5 h), 25 (100, 300, 0.5 h, and
100, 300, 4 h), 31 (100, 300, 0.5 h), 35 (300, 4 h), 39 (30, 100,
300, 0.5 h, and 100, 300, 4 h), 40 (300, 0.5 h, and 100, 300, 4
h), 41 (300, 0.5 and 4 h), 42 (100, 300, 4 h), 48 (100, 300, 0.5
h, 4 h), 50 (30, 0.5 h, and 300, 4 h), 51 (100, 300, 4 h), 52 (100,
300, 4 h), 53 (30, 100, 300, 0.5 h, and 100, 300, 4 h), 54 (30,
100, 0.5 h, and 100, 300, 4 h), 55 (100, 0.5 h), 56 (300, 0.5 h,
and 100, 4 h), 57 (100, 300, 0.5 h), 71 (300, 0.5 h), 76 (300, 0.5
h), 78 (300, 0.5 h, 4 h), 79 (300, 4 h for three of five mice), and
84 (300, 0.5 h, 4 h). Myoclonic jerks were observed in the
following compounds, namely 9 (100, 300, 0.5 h), 14 (30, 100,
0.5 h), 42 (300, 0.5 h), 55 (300, 0.5 h), 56 (100, 0.5 h), 71
(100, 0.5 h for one of five mice), 79 (300, 4 h for two of five
mice). Death following CSA was caused by the following
compounds, namely 22 (300, 4 h), 56 (300, 4 h), 76 (100, 0.5 h,
and 300, 4 h). Tonic extension was observed during the
evaluation of 33 (30, 0.5 h). The mice died during the test
without having seizure in the case of 49 (300, 0.5 h), 71 (300,
4 h), 86 (100, 300, 0.5 h), and 87 (100, 0.5 h for one of five
mice, 300, 0.5 h for all mice and 30, 4 h for two of five mice).
Stretching and rolling were noted for mice caused by 75 (300,
4 h), and tremors resulted from the administration of 85 (300,
4 h). Death occurred after injections of 86 (100, 300, 4 h) and
87 (100, 300, 4 h).
In the NT screen, mice were unable to grasp the rotorod
after administration of the following compounds, viz. 9 (300,
4 h), 13 (300, 4 h), 14 (100, 300, 4 h), 16 (100, 300, 4 h), 18
(100, 300, 4 h), 19 (300, 4 h in one of two mice), 22 (100, 300,
0.5 h, and 300, 4 h), 25 (100, 300, 4 h), 30 (300, 4 h), 32 (300,
4 h), 35 (300, 4 h), 39 (300, 0.5 h, and 100, 300, 4 h), 40 (300,
0.5 and 4 h), 42 (300, 4 h), 49 (300, 0.5 h, 4 h), 50 (300, 0.5 h,
4 h), 51 (300, 4 h), 52 (300, 4 h), 53 (300, 0.5 h, and 100, 300,
4 h), 54 (300, 4 h), 55 (300, 4 h), 56 (100, 300, 4 h), 59 (300,
0.5 h), 71 (300, 4 h), 75 (300, 4 h), 76 (300, 0.5 h, and 300, 4
h), 78 (300, 4 h), 84 (300, 4 h), 85 (300, 4 h), 86 (300, 0.5 h),
and 90 (300, 4 h). A loss of righting reflex was noted with 87
(300, 0.5 h).
Ack n ow led gm en t. The authors thank Nordic Mer-
rell Dow Research, Laval, PQ, Canada, for financial
support of this project and the Antiepileptic Drug
Development Program, NIH, for generating the biologi-
cal data. The Natural Sciences and Engineering Re-
search Council of Canada provided operating and equip-
ment grants to J . W. Quail and a Chemistry Under-
graduate Student Research Award to T. Lechler. J . M.
Smith was the recipient of a Pharmaceutical Manufac-
turers Association of Canada-Health Research Founda-
tion/Medical Research Council of Canada Summer
Research Award. Appreciation is recorded to L. Prasad
who gave assistance in the interpretation of the X-ray
crystallographic data and also with the use of the
MacroModel program. Mrs. Z. Dziadyk and Mrs. S.
Thiessen are thanked for typing the various drafts of
the manuscript.
Su p p or tin g In for m a tion Ava ila ble: Atomic anisotropic
displacement parameters, hydrogen positional and isotropic
displacement parameters, atomic positional and equivalent
isotropic displacement parameters, and bond distances and
angles for 1, 2, 5, 23, and 73 and tables of the measurements
indicated in Figure 9 (19 pages). Ordering information is given
on any current masthead page.
Refer en ces
(1) Dimmock, J . R.; Sidhu, K. K.; Thayer, R. S.; Mack, P.; Duffy, M.
J .; Reid, R. S.; Quail, J . W.; Pugazhenthi, U.; Ong, A.; Bikker,
J . A.; Weaver, D. F. Anticonvulsant activities of some arylsemi-
carbazones displaying potent oral activity in the maximal
electroshock screen in rats accompanied by high protection
indices. J . Med. Chem. 1993, 36, 2243-2252.
(2) 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.
(3) Dimmock, J . R.; Sidhu, K. K.; Tumber, S. D.; Basran, S. K.; Chen,
M.; Quail, J . W.; Yang, J .; Rozas, I.; Weaver, D. F. Some aryl
semicarbazones possessing anticonvulsant activities. Eur. J .
Med. Chem. 1995, 30, 287-301.
(4) Dimmock, J . R.; Pandeya, S. N.; Quail, J . W.; Pugazhenthi, U.;
Allen, T. M.; Kao, G. Y.; Balzarini, J .; De Clercq, E. Evaluation
of the semicarbazones, thiosemicarbazones and bis-carbohydra-
zones of some aryl alicyclic ketones for anticonvulsant and other
biological properties. Eur. J . Med. Chem. 1995, 30, 303-314.
(5) Silverman, R. B. The Organic Chemistry of Drug Design and
Drug Action; Academic Press, Inc.: San Diego, 1992; p 28.
(6) Albert, A. Selective Toxicity, 7th ed.; Chapman and Hall:
London, 1985; pp 611-624.
The quantitative evaluation of the effect of selected com-
pounds after intraperitoneal injection in mice in the MES,
scPTZ, and NT screens is summarized in Table 2.
(ii) Or a l Ad m in istr a tion to Ra ts. Most of the compounds
described in this study were examined for oral activity in the
MES and NT screens. The results are presented in Table 1.
In addition, a number of semicarbazones were evaluated in
the scPTZ screen using the doses specified in Table 1 (except
for 27 and 42 vide infra), and the results are as follows.
Compounds 4, 7, 10, 13, 26, 32, 71, 84, and 85 were devoid of
activity in the scPTZ screen. On the other hand, the following
compounds had activity (number of rats protected out of 4,

(Aryloxy)aryl Semicarbazones and Related Compounds
J ournal of Medicinal Chemistry, 1996, Vol. 39, No. 20 3997
(7) Dimmock, J . R.; J onnalagadda, S. S.; Hussein, S.; Tewari, S.;
Quail, J . W.; Reid, R. S.; Delbaere, L. T. J .; Prasad, L. Evaluation
of some thiosemicarbazones of arylidene ketones and analogues
for anticonvulsant activities. Eur. J . Med. Chem. 1990, 25, 581-
588.
(8) Lock, G.; Kempter, F. H. Derivatives of phenyl ether. II.
Monoaldehydes. Monatsh. Chem. 1935, 67, 24-35.
(9) Tomita, M.; Edatsune, S.; Iwata, S. Synthesis of aldehyde
derivatives containing a diphenyl ether nucleus. J . Pharm. Soc.
J pn. 1955, 75, 1021-1023.
(10) Manske, H. F.; Ledingham, A. E. Synthesis and reactions of some
dibenzoxepines. J . Am. Chem. Soc. 1950, 72, 4797-4799.
(11) Dimmock, J . R.; McColl, J . M.; Wonko, S. L.; Thayer, R. S.;
Hancock, D. S. Evaluation of the thiosemicarbazones of some
aryl alkyl ketones and related compounds for anticonvulsant
activities. Eur. J . Med. Chem. 1991, 26, 529-534.
(12) Dimmock, J . R.; Puthucode, R. N.; Lo, M. S.; Quail, J . W.; Yang,
J .; Stables, J . P. Structural modifications of the primary amino
group of anticonvulsant aryl semicarbazones. Pharmazie 1996,
51, 83-88.
(14) Hansch, C.; Leo, A. J . Substituent Constants for Correlation
Analysis in Chemistry and Biology; J ohn Wiley and Sons, Inc.:
New York, 1979; pp 49-51, 115, 170-172.
(15) Taft, R. W., J r. Separation of Polar, Steric and Resonance Effects
in Reactivity. In Steric Effects in Organic Chemistry; Newman,
M. S., Ed.; J ohn Wiley and Sons, Inc.: New York, 1956; p 591.
(16) MacroModel Version 4.5, Department of Chemistry, Columbia
University, New York, August 1994.
(17) Mohamadi, F.; Richards, N. G. J .; Guide, W. C.; Liskamp, M.
L.; Caufield, C.; Chang, G.; Hendrikson, T.; Still, W. C. Macro-
Model - An integrated software system for modeling organic and
bioorganic molecules using molecular mechanics. J . Comput.
Chem. 1990, 11, 440-467.
(18) Gabe, E. J .; LePage, Y.; Charland, J . P.; Lee, F. L.; White, P. S.
An interactive program system for structure analyses. J . Appl.
Crystallogr. 1989, 22, 384-387.
(19) International Tables for X-ray crystallography; Kynoch Press:
Birmingham, 1974; Vol. IV.
(20) Porter, R. J .; Cereghino, J . J .; Gladding, G. D.; Hessie, B. J .;
Kupferberg, H. J .; Scoville, B.; White, B. G. Antiepileptic drug
development program. Cleveland Clin. Q 1984, 51, 293-305.
(13) Dimmock, J . R.; Phillips, O. A.; Wonko, S. L.; Hickie, R. A.; Tuer,
R. G.; Ambrose, S. J .; Reid, R. S.; Mutus, B.; Talpas, C. J .
Evaluation of some Mannich bases of conjugated styryl ketones
and related compounds versus the WiDr colon cancer in vitro.
Eur. J . Med. Chem. 1989, 24, 217-226.
J M9603025