Multitargeted Compounds Derived from (2,5-Dioxopyrrolidin-1-yl)(phenyl)-Acetamides as Candidates for Effective Anticonvulsant and Antinociceptive Agents

Article abstract of DOI:10.1021/acschemneuro.0c00257

We developed a focused set of original hybrid pyrrolidine-2,5-dione derivatives with potent anticonvulsant and antinociceptive properties. These hybrid compounds demonstrated broad-spectrum protective activity in a range of mouse models, such as the maximal electroshock (MES) test, the pentylenetetrazole-induced seizures (scPTZ), and the 6 Hz (32 mA) seizures. Compound 22 showed the most potent anticonvulsant activity (ED50 MES = 23.7 mg/kg, ED50 6 Hz (32 mA) = 22.4 mg/kg, ED50 scPTZ = 59.4 mg/kg). In addition, 22 revealed potent efficacy in the formalin-induced tonic pain. These in vivo activities of 22 are likely mediated by several targets and may result from the inhibition of central sodium/calcium currents and transient receptor potential vanilloid 1 (TRPV1) receptor antagonism. Finally, the lead compound 22 revealed drug-like absorption, distribution, metabolism, excretion, toxicity (ADME-Tox) properties in the in vitro assays, making it a potential candidate for further development in epilepsy and neuropathic pain indications.

Full text of DOI:10.1021/acschemneuro.0c00257

Subscriber access provided by Uppsala universitetsbibliotek
Article
Multitargeted Compounds Derived from (2,5-
Dioxopyrrolidin-1-yl)(phenyl)-Acetamides as Candidates
for Effective Anticonvulsant and Antinociceptive Agents
Micha# Abram, Anna Rapacz, Szczepan Mogilski, Gniewomir Latacz,
Annamaria Lubelska, Rafa# M. Kami#ski, and Krzysztof Kami#ski
ACS Chem. Neurosci., Just Accepted Manuscript • DOI: 10.1021/acschemneuro.0c00257 • Publication Date (Web): 01 Jun 2020
Downloaded from pubs.acs.org on June 2, 2020
Just Accepted
“Just Accepted” manuscripts have been peer-reviewed and accepted for publication. They are posted
online prior to technical editing, formatting for publication and author proofing. The American Chemical
Society provides “Just Accepted” as a service to the research community to expedite the dissemination
of scientific material as soon as possible after acceptance. “Just Accepted” manuscripts appear in
full in PDF format accompanied by an HTML abstract. “Just Accepted” manuscripts have been fully
peer reviewed, but should not be considered the official version of record. They are citable by the
Digital Object Identifier (DOI®). “Just Accepted” is an optional service offered to authors. Therefore,
the “Just Accepted” Web site may not include all articles that will be published in the journal. After
a manuscript is technically edited and formatted, it will be removed from the “Just Accepted” Web
site and published as an ASAP article. Note that technical editing may introduce minor changes
to the manuscript text and/or graphics which could affect content, and all legal disclaimers and
ethical guidelines that apply to the journal pertain. ACS cannot be held responsible for errors or
consequences arising from the use of information contained in these “Just Accepted” manuscripts.
is published by the American Chemical Society. 1155 Sixteenth Street N.W.,
Washington, DC 20036
Published by American Chemical Society. Copyright © American Chemical Society.
However, no copyright claim is made to original U.S. Government works, or works
produced by employees of any Commonwealth realm Crown government in the course
of their duties.

Page 1 of 38
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
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
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 2 of 38
1
2
3
4
5
6
7
8
9
Multitargeted Compounds Derived from (2,5-Dioxopyrrolidin-1-yl)(phenyl)-
Acetamides as Candidates for Effective Anticonvulsant and Antinociceptive
Agents
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
Michał Abram †, Anna Rapacz #, Szczepan Mogilski #, Gniewomir Latacz ‡, Annamaria
Lubelska ‡, Rafał M. Kamiński †, and Krzysztof Kamiński †,*
† Jagiellonian University Medical College, Faculty of Pharmacy, Department of Medicinal
Chemistry, Medyczna 9, 30-688 Cracow, Poland
# Jagiellonian University Medical College, Faculty of Pharmacy, Department of
Pharmacodynamics, Medyczna 9, 30-688 Cracow, Poland
‡ Jagiellonian University Medical College, Faculty of Pharmacy, Department of Technology
and Biotechnology of Drugs, Medyczna 9, 30-688 Cracow, Poland
*Corresponding Author:
K.K.: Phone, +48 12 620 54 59; fax, +48 12 620 54 58; e-mail, k.kaminski@uj.edu.pl.
ACS Paragon Plus Environment

Page 3 of 38
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
ABSTRACT: We developed a focused set of original hybrid pyrrolidine-2,5-dione derivatives
with potent anticonvulsant and antinociceptive properties. These hybrid compounds
demonstrated broad-spectrum protective activity in a range of mouse models, such as the
maximal electroshock (MES) test, the pentylenetetrazole-induced seizures (scPTZ), and the 6
Hz (32 mA) seizures. Compound 22 showed the most potent anticonvulsant activity (ED50 MES
= 23.7 mg/kg, ED50 6 Hz (32 mA) = 22.4 mg/kg, ED50 scPTZ = 59.4 mg/kg). In addition, 22
revealed potent efficacy in the formalin-induced tonic pain. These in vivo activities of 22 are
likely mediated by several targets and may result from inhibition of central sodium/calcium
currents and TRPV1 receptor antagonism. Finally, the lead compound 22 revealed drug-like
ADME-Tox properties in the in vitro assays making it a potential candidate for further
development in epilepsy and neuropathic pain indications.
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
KEYWORDS: Hybrid compounds, multi-targeted drugs, epilepsy, neuropathic pain,
anticonvulsants, antinociceptive activity.
INTRODUCTION
A considerable interest in designing new multi-targeted compounds has been observed
in recent years. Such compounds have been proven to be advantageous in the treatment of
multifactorial diseases (characterized by complex pathomechanisms) often linked with drug
resistance.1–3 Epilepsy, which is recognized as a common neurological disorder clearly falls into
this category. In fact, about one-third of the patients with epilepsy do not respond to current
pharmacological treatments and are diagnosed with drug-resistant epilepsy.4 Multifunctional
(or multi-targeted) antiepileptic drugs (AEDs), such as valproic acid (VPA), belong to the most
frequently prescribed and most effective compounds in different types of epilepsies.5,6 The
combination of different mechanisms of action tends to be also more beneficial in case of drug-
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 4 of 38
1
2
3
4
5
6
7
8
9
resistant epilepsy.7,8 Therefore multi-targeted AEDs, which involve synergistic or additive
mechanisms, may provide a more comprehensive spectrum of activity compared to single-
targeted drugs as well as may be characterized by a better therapeutic window. This is
exemplified by padsevonil, a dual acting AED candidate that integrates pharmacophoric
features of both GABAA (benzodiazepine receptor, BZD) and SV2A (synaptic vesicle
glycoprotein 2A) ligands.9 Consequently, padsevonil displays both a broad profile of
anticonvulsant activity and an improved therapeutic window versus many other AEDs.10
In keeping with the above-described concepts of multi-targeted drugs and with the aim
of obtaining new highly effective broad-spectrum anticonvulsants, we have developed a series
of integrated hybrid molecules derived from the pyrrolidine-2,5-dione ring.11–13 These
compounds were designed by applying the fragment-based approach14 incorporating common
structural features found in three chemically and pharmacologically diverse AEDs, such as
ethosuximide (ETX, active primarily in the pentylenetetrazole-induced seizure model (scPTZ)),
levetiracetam (LEV, active in the 6 Hz (32 mA) model), and lacosamide (LCS, active in both
maximal electroshock (MES) and 6 Hz (32/44 mA) seizure models). As a result, we obtained
hybrid compounds with potent and broad-spectrum anticonvulsant activity sharing the
pharmacological properties of all aforementioned AEDs. The most beneficial anticonvulsant
and safety profile was observed for compound KA-11 (Figure 1). In addition to extensive
anticonvulsant activity of KA-11 (MES, scPTZ, and 6 Hz seizure models), this substance was
also found to effectively decrease pain responses in the formalin-induced tonic pain, the
capsaicin-induced neurogenic pain, and notably, in the oxaliplatin-induced neuropathic pain in
mice.10,11 Aiming to further improve the anticonvulsant activity of KA-11, particularly in the
MES and 6 Hz (32 mA), we now developed a new series of (2,5-dioxopyrrolidin-1-
yl)(phenyl)acetamides (Figure 1A). We hypothesized that the presence of an additional
aromatic ring (marked in blue) in the vicinity of the amide moiety may provide an orientation
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
ACS Paragon Plus Environment

Page 5 of 38
ACS Chemical Neuroscience
1
2
3
4
5
6
that is more comparable to that observed in the structure of lacosamide, and thus will improve
protection in the MES and 6 Hz (32 mA) models in mice.
7
8
9
The transient receptor potential vanilloid 1 (TRPV1) is a nonselective cation channel,
expressed mainly in the afferent sensory neurons, where it plays a key role in the detection of
noxious painful stimuli. Thus, TRPV1 antagonists are recognized primarily as potential drugs
useful in the treatment of different pain syndromes, including neuropathic pain, pain associated
with inflammation and migraine, but also epilepsy, diabetes, cough, bladder disorders, and
hearing loss.15,16 Bearing in mind the aforementioned facts, the introduction of phenyl
substituent (marked in blue) also brought the structure of these new compounds closer to that
observed in known TRPV1 antagonists e.g. piperazine carboxamide derivative – BCTC (Figure
1), and in consequence may lead to an improvement in their antinociceptive activity. We also
aimed to assess the influence of certain structural modifications on anticonvulsant activity by
further expansion of structural diversity, i.e. the distance between aromatic side ring and the
acetamide linker (Figure 1B, marked in orange), the kind and size of the heterocyclic amine
(e.g. piperazine, piperidine or diazepine, Figure 1C, marked in yellow), and finally the presence
of the benzene ring as rigidified fragment of the molecule (Figure 1D, marked in green). In
addition, we have assessed several ADME-Tox parameters in the in vitro assays as well as we
attempted to elucidate the possible mechanism of action.
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
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 6 of 38
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
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
Figure 1. Design strategy and general structure of new hybrid molecules.
ACS Paragon Plus Environment

Page 7 of 38
ACS Chemical Neuroscience
1
2
3
4
RESULTS AND DISCUSSION
5
6
7
8
Chemistry. The (monosubstituted)phenylpiperazinamides - compounds 11–41 (Series
1) were obtained by using a multi-step procedure according to Scheme 1. First, we synthesized
succinamic acids (1–5) by reacting equimolar amounts of commercially available succinic
anhydride with DL-phenylglycine (or its fluorinated analogs) or DL-phenylalanine. Next, the
hexamethyldisilazane (HMDS)-promoted cyclization reaction of 1–5 was performed according
to a previously reported method,9 yielding intermediate monocarboxylic acids 6–10. The final
9
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
compounds
11–41
were
obtained
by
coupling
6–10
with
diverse
(monosubstituted)phenylpiperazines (containing both electron-withdrawing or electron-
donating atom/group) in the presence of carbonyldiimidazole (CDI) as the coupling reagent.
This reaction was performed at room temperature in dry dimethylformamide (DMF), and its
progress was monitored using the high-performance liquid chromatography (HPLC)
(completion at ~24 h).
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 8 of 38
1
2
3
O
O
CH
3COOH
4
5
6
7
8
9
70o
O
12
h
C,
HO
O
O
1: R
2: R
3: R
4: R
5: R
=H, (n=0)
=H, (n=1)
=3-F, (n=0)
=4-F, (n=0)
HN
OH
O
n
(CH2)
O
H2N
=3,4-diF, (n=0)
OH
R
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
n
(CH2)
1-5
R
HMDS, ZnCl2
Benzene, reflux, 12 h
O
6: R
O
=H, (n=0)
=H, (n=1)
=3-F, (n=0)
=4-F, (n=0)
7: R
N
8:
9: R
OH
R
O
n
(CH2)
10: R
=3,4-diF, (n=0)
R
6-10
HN
CDI, DMF
r.t., 24 h
N
R1
O
O
N
N
O
N
n
(CH2)
R1
R
37-41
11-36
(n=0),
(n=1)
= H, 1= H
= H, 1= 2-Cl
= H, 1= 3-Cl
= H, 1= 4-Cl
= H, 1= 2-F
= H, 1= 3-F
= H, 1= 4-F
= H, 1= 2-CH3
= H, 1= 3-CH3
= H, 1= 4-CH3
= H, 1= 2-CF3
= H, 1= 3-CF3
= H, 1= 3-OCH3
R
11: R
12: R
13: R
14: R
15: R
16: R
17: R
18: R
19: R
20: R
21: R
22: R
23: R
R
24: R
25: R
26: R
27: R
28: R
29: R
30: R
31: R
32: R
33: R
34: R
35: R
36: R
= H, 1= H
37: R
R
R
= H, 1= 3-OCF3
R
= H, 1= 3-Cl
= H, 1= 3-CH3
R
38: R
39: R
40: R
41: R
R
R
= H, 1= 3-SCH3
R
= H, 1= 3-SCF3
R
= H, 1= 3-CF3
R
R
R
= H, 1= 3-CHF2
R
= H, 1= 3-OCF3
R
R
= H, 1= 3-NH2
R
R
= H, 1= 3-CH(CH3 2
R
)
R
= H, 1= 3-C(CH3 3
R
)
R
= H, 1= 3-C
R
5H6
= H, 1= 3-O-CH
R
R
H6
2C5
= 3-F, 1= 3-CF3
R
R
= 4-F, 1= 3-CF3
R
R
= H, 1= 4-CF3
= 3,4-diF, 1= 3-CF3
R
R
Scheme 1. Synthesis of intermediates 1–10 and target compounds 11–41 (Series 1).
ACS Paragon Plus Environment

Page 9 of 38
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
Based on the in vivo anticonvulsant activity data obtained for compounds representing
Series 1, in the aim of finding the optimal substitution pattern of the phenylpiperazine moiety,
the following structural modifications were focused on substances containing two electron
withdrawing atoms/groups at the aromatic ring (Series 2). In this case we tested at first different
position of chlorine atoms as follow 2,3-diCl (42), 3,4-diCl (43), and 3,5-diCl (44); notably for
each compound we retained one chlorine at position-3 that appeared to be especially preferable
for the anticonvulsant activity of the monochloro-derivative 13 (see Table S1 and Table 1). As
a result, the most potent anticonvulsant activity was noticed for 3,5-diCl derivative 44, therefore
the substitution of other molecules was restricted only to position 3 and 5. The structures and
synthetic pathway for compounds of Series 2 are depicted on Scheme 2.
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
O
O
1= 2-Cl, 2= 3-Cl
1= 3-Cl, 2= 4-Cl
42: R
43: R
44: R
45: R
46: R
47: R
R
O
O
CDI, DMF
r.t., 24 h
R
N
N
N
OH
1= 3-Cl, 2= 5-Cl
R
O
O
N
1= 3-CF3 2= 5-CF3
R
,
R1
1= 3-SCF3 2= 5-SCF3
R
,
1= 3-Cl, 2= 5-CF3
R
R2
HN
6
42-47
N
R1
R2
Scheme 2. Synthesis of target compounds 42–47 (Series 2).
Finally, with the aim of examining the influence of the piperazine moiety on
anticonvulsant activity for the most potent anticonvulsant from Series 1 that is, compound 22,
the respective piperidine (48) and diazepine (49) analogs were synthesized. Furthermore, for
compound 22 we obtained also two derivatives 52 and 53 with the rigid benzene ring (Scheme
3). The latter compounds were synthesized according to the two-step procedure. Thus, the
cyclocondensation reaction of succinic anhydride with 3-amino- or 4-aminobenzoic acid
yielded corresponding intermediates 48 and 49, which were then coupled with 1-[3-(trifluoro-
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 10 of 38
1
2
3
4
5
6
7
8
methyl)phenyl]piperazine in the presence of CDI to form the desired compounds 50 and 51.
The structures and synthetic routs for intermediates 50, 51 and target compounds 48, 49, 52,
and 53 (Series 3) are depicted in Scheme 3.
9
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
O
O
O
O
CDI, DMF
r.t., 24 h
m
X
X
R
R
N
1= 3-CF3
1= 3-CF3
N
48: m
49: m
=1,
=C,
N
OH
=2, =N,
O
O
X
R1
m
X
HN
6
48 49
,
R1
O
O
N
N
HN
O
N
CF3
H2N
N
CF3
COOH
52
O
O
CDI, DMF
r.t., 24 h
N
O
O
180 o
C, 1 h
COOH
O
50:
51:
O
m-COOH
p-COOH
N
N
CF3
O
N
O
53
Scheme 3. Synthesis of target compounds 48, 49, 52, and 53 (Series 3).
In case of compounds 25–28, 30–33, and 44–47 the staring 4-phenylpiperazines (A13–
A24) were previously synthesized using the respective aryl bromides and 1-Boc-piperazine in
the Buchwald–Hartwig amination reaction in the nitrogen atmosphere.17 Removal of the tert-
butoxycarbonyl group (Boc) with trifluoroacetic acid (TFA) followed by the neutralization with
25% ammonium hydroxide produced the desired 4-phenylpiperazine derivatives A13–A24
(Scheme 4).
ACS Paragon Plus Environment

Page 11 of 38
ACS Chemical Neuroscience
1
2
3
4
5
,
Pd2dba3 BINAP,
Sodium tert-butoxide,
Toluene, reflux, 12 h
1) TFA, r.t., 5 h
R1
R1
O
O
O
O
4OH
2) 25% NH
HN
N
N
NH
N
N
6
R2
R2
7
8
9
R1
A1-A12
A13-A24
Br
3-CH(CH
H
3-CH(CH
H
=
=
A13: R1
R2
=
H
=
=
=
H
A1: R1
A2: R1
A3: R1
A4: R1
A5: R1
A6: R1
A7: R1
A8: R1
A9: R1
R2
=
H
,
)
3 2
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
R2
,
)
,
3 2
R2
H
A14: R1
A15: R1
A16: R1
A17: R1
A18: R1
A19: R1
A20: R1
A21: R1
A22: R1
A23: R1
A24: R1
R2
=
=
3-C(CH3 3
)
,
3-C(CH3 3
)
=
R2
R2
=
R2
R2
,
=
=
3-OCF3
3-OCF3
,
H
3-C
H
=
=
=
=
3-C H6
,
H6
,
3-O5-CH
H
3-O5-CH
3-SCH
3-SCF3
H
H
=
R2
=
=
=
=
=
=
R2
,
H6
2C5
R2
R2
,
R2
H6
2C5
R2
R2
R2
,
3-SCH
H
H
H
H
H
H
=
=
=
=
=
=
=
=
=
,
,
3-SCF3
=
=
,
3-CHF3
3-CHF3
=
=
,
,
2
R2
2
R2
3-Cl,
5-Cl
3-Cl,
5-Cl,
=
=
=
R2
5-CF
=
3-CF
3-SC3F
5-CF
3-CF
=
R2
R2
=
=
=
A10: R1
A11: R1
A12: R1
R2
,
,
3
3
3-SC3F
5-SCF
5-SCF
=
=
R2
,
,
3
R2
3
3
3
3-Cl,
5-CF
3-Cl,
5-CF
=
=
R2
3
3
Scheme 4. Synthesis of starting 4-phenylpiperazine derivatives A13–A24.
The target hybrid compounds were obtained in satisfying yield (>60%). The structures
of initial 4-phenylpiperazine derivatives, carboxylic intermediates and desired molecules were
1
13
19
confirmed by H NMR, C NMR (selected molecules), F NMR (selected fluorinated final
substances), and/or LC-MS spectra. The purity of target substances determined by UPLC
method was ≥95%. The details are summarized in the Materials and Methods section and also
Supporting Information.
Anticonvulsant activity. The screening approach applying animal models of epilepsy
is an important and efficient pharmacological tool for the discovery of new anticonvulsants, as
it enables the identification of both compounds with well-established mechanisms of action
among known AEDs as well as compounds with novel and previously unidentified
pharmacodynamics.18 Therefore, the anticonvulsant properties of all final compounds was
initially assessed using the MES test, which is a recognized model of tonic–clonic epilepsy in
human.19 This test allows identification of compounds able to prevent the spread of seizures
when all neuronal circuits in the brain are paroxysmally active. It is worth noting that despite
significant advances in epilepsy research during the past few decades, MES test is still
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 12 of 38
1
2
3
4
5
6
7
8
9
recognized as the most useful preclinical seizure model. It is widely employed during the early
identification and screening of new AED candidates, as it may help to identify compounds with
unique mechanisms of action.20 In this study, all final compounds were initially tested after
intraperitoneal (i.p.) administration at a fixed dose - 100 mg/kg in mice (in groups consisting
of four animals) and the protection against MES seizures was observed at two pretreatment time
points of 0.5 and 2 h.
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
According to the results of the MES test obtained for the (monosubstituted)-
phenylpiperazinamides 11–41 (Series 1), at least 75% protection (3 mice protected from 4
tested) was demonstrated by compounds 11, 13, 19, 20, 22, 23, 25, 27, 28, 32, and 35, which
was found to be satisfactory (Table S1). No lethality was observed for these substances
administrated at dose of 100 mg/kg. All aforementioned compounds were found to be active
predominantly after a short pre-administration time (0.5 h), whereas compounds 19, 22, 25, 27
and 32 were found to be active at both time points (0.5 and 2 h). Notably, 3-CF3 (22), 3-OCF3
(25), and 3-SCF3 (27) derivatives provided maximal (100%) protection at 0.5 and 2 h. In the
next step of the pharmacological characterization, eleven aforementioned MES-active
compounds were studied in the mouse 6 Hz (32 mA) model of seizures, which is a well-
established and commonly used preclinical model in the discovery of new AEDs effective in
human focal epilepsy (Table S1).18 According to the results, majority derivatives (11, 13, 19,
22, 23, 25, 27, 28, 32, and 35) displayed satisfactory anticonvulsant activity protecting at least
75% of the mice especially at the time point of 0.5 h (only compound 13 showed potent
protection at both pretreatment times). The maximal (100%) protection was observed for 3-Cl
(13), 3-CH3 (19), 3-CF3 (22), 3-OCF3 (25), 3-SCF3 (27), and 3-phenyl (32) congeners. With
the aim of finding broad-spectrum anticonvulsants during the preclinical tests, all effective
compounds in the MES and 6 Hz (32 mA) models (11, 13, 19, 22, 23, 25, 27, 28, 32, and 35)
were further studied using the scPTZ test. In the latter test seizures are induced chemically and
ACS Paragon Plus Environment

Page 13 of 38
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
the latency and incidence of clonic seizures are the primary endpoints (Table S2). The scPTZ
model tends to identify drugs that may be clinically effective against generalized non-
convulsive (myoclonic, absence) seizures.21 In this test, compound 22 was most active and
displayed 100% and 75% protection at 0.5 and 2 h, respectively. A satisfactory efficacy (75%)
was also observed for compounds 11, 13, 19, and 28 at 0.5 h. Surprisingly, the in vivo data for
disubstituted piperazinamides 42–47 (Series 2) showed beneficial anticonvulsant properties
only for 3,5-dichloroderivative (44), which protected mice in all tests applied; MES, 6 Hz (32
mA) as well as scPTZ. During further structure-activity relationship analysis (SAR) of the
aforementioned derivatives, we investigated the influence of the distance between the side
aromatic ring and the acetamide linker on their anticonvulsant properties (modification B,
Figure 1). Thus, structural modification of the selected potent phenylglycinamides (11, 13, 19,
22, and 25) resulted in the production of respective phenylalaninamides (37–41, see Scheme 1),
which were subjected to the MES test. According to the results (Table S1), elongation of the
distance between the aforementioned structural fragments (acetamide and side aromatic ring)
caused a decrease in the anticonvulsant activity, which yielded compounds with efficacy equal
to 50%. Regarding the influence of heterocyclic amine on the anticonvulsant activity
(modification C, Figure 1), the presence of piperazine moiety proved to be especially beneficial,
as we observed 75% protection in the MES and 6 Hz (32 mA) seizure models for piperidine
analog 48 or lack of activity in case of diazepine derivative 49 (Table S3). Similar observations
were noted for analogs with rigidified benzene ring 48 (meta-isomer) and 49 (para-isomer),
which may be regarded as the stiffened analogs of the most potent compound 22 (Table S3,
modification D, Figure 1).
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
In summary, based on the above screening data, (2,5-dioxopyrrolidin-1-
yl)(phenyl)acetamides with unsubstituted phenylpiperazine (11) and respective 3-Cl (13), 3-
CH3, (19), 3-CF3 (22), 3-CHF2 (28), and 44 (3,5-diCl) analogs revealed broad-spectrum
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 14 of 38
1
2
3
4
5
6
7
8
9
anticonvulsant activity and potent protection in the MES, scPTZ, and 6 Hz (32 mA) seizure
models. The replacement of the methyl or trifluoromethyl groups onto para-position (20, 23),
the introduction of trifluoromethoxy (25), trifluorotiomethyl (27), and phenyl (32) moieties at
meta-position, the introduction of fluorine atom at para-position of the side aromatic ring (35),
or exchange of the piperazine ring to piperidine moiety (48) yielded compounds that are active
predominantly in the MES and 6 Hz (32 mA) tests. All other structural modifications such as
the other modes of substitution of both aromatic areas, insertion of the benzyl moiety at the
acetamide linker, insertion of the diazepine in place of piperazine moiety, or the presence of
benzene ring as the rigid fragment of molecules decreased the anticonvulsant activity or
rendered the compounds inactive.
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
Continuing the preclinical characterization of the aforementioned hybrids, for
compounds that showed at least of 75% protection in preliminary studies, their median effective
doses (ED50) in the given seizure model and median doses causing neurological impairment
(TD50) in the rotarod test were determined. These two pharmacological parameters were used
to calculate the protective indexes (PIs), which are a measure of potential therapeutic window
of the tested agent. In aim of assessing the potential for clinical utility of the compounds
obtained herein, we performed the same experiments using AEDs with a narrow spectrum of
efficacy in the preclinical studies (LEV – effective in 6 Hz, 32 mA test; ETX – effective in
scPTZ test); anticonvulsants with wider activity (LCS – effective in MES and 6 Hz (32/44
mA) tests); and with VPA, which is recognized as a broad-spectrum AED (active in MES,
scPTZ, and 6 Hz (32/44 mA) seizure models). Notably, VPA possesses multiple mechanisms
of action and has efficacy against the widest range of epilepsy types and multiple epileptic
syndromes. Table 1 summarizes the results obtained.
ACS Paragon Plus Environment

Page 15 of 38
ACS Chemical Neuroscience
1
2
3
4
Table 1. The quantitative pharmacological parameters ED50, TD50, and PIs in mice i.p.
ED50 6 Hz
5
6
7
8
Compd TPE
(h)a
ED50 MES
(mg/kg)b
ED50 scPTZ
(mg/kg)d
TD50
(mg/kg)e
PI
(32 mA)
(mg/kg)c
(TD50/ED50)f
9
>3.3 (MES)
>3.6 (6 Hz)
>3.0 (scPTZ)
3.9 (MES)
3.6 (6 Hz)
2.7 (scPTZ)
4.6 (MES)
4.8 (6 Hz)
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
91.1
(84.5–98.5)
83.5
(65.9–105.7)
100.0
(87.3–114.5)
11
13
19
22
0.5
0.5
0.5
0.5
>300
36.9
(32.2–42.4)
39.5
(28.3–55.2)
52.6
(37.7–73.5)
143.8
(113.8–181.9)
37.2
(33.2–41.8)
35.5
(25.6–49.4)
57.6
(34.0–97.9)
171.0
(157.2–185.9)
3.0 (scPTZ)
8.2 (MES)
8.7 (6 Hz)
23.7
(18.4–31.2)
22.4
(17.4–28.8)
59.4
(37.5–94.1)
195.7
(132.7–288.6)
3.3 (scPTZ)
94.3
(85.0–104.8)
23
32
27
0.5
0.5
0.5
>100
–
–
–
>300
>3.2 (MES)
27.4
(23.0-32.5)
14.1
(14.2–48.5)
66.2
(52.6–83.2)
2.4 (MES)
4.7 (6 Hz)
36.2
(31.3–41.8)
15.8
(7.05–19.8)
150.1
(133.9–168.2)
4.1 (MES)
9.5 (6 Hz)
2.8 (MES)
4.4 (6 Hz)
2.9 (scPTZ)
97.8
(82.0–116.6)
63.0
(59.7–66.5)
94.4
(85.0–104.8)
274.2
(257.8–291.7)
28
0.5
43.9
(40.8–47.3)
26.2
(8.4–23.5)
75.6
(65.1–87.8)
1.7 (MES)
2.9 (6 Hz)
32
0.5
0.5
0.5
0.5
0.25
0.5
1.0
–
–
56.4
(39.9–79.8)
48.3
(43.1–54.1)
5.3 (MES)
6.2 (6 Hz)
35
>300
>300
68.5
(62.1–75.6)
17.7
(13.7–22.8)
>4.4 (MES)
>17.0 (6 Hz)
44
>100
81.8
(76.3–87.7)
41.1
(27.2–62.0)
254.3
(221.2–292.4)
3.1 (MES)
6.2 (6 Hz)
48
–
140.4
318.0
ETXg
LCSg
LEVg
n.a.
>200
2.3 (scPTZ)
(115.8–170.2) (295.8–341.9)
9.2
(8.5–10.0)
5.3
(3.5–7.8)
46.2
n.a.
5.0 (MES)
8.8 (6 Hz)
(44.5–48.0)
15.7
(10.4–23.7)
>500
n.a.
>500
>31.8 (6 Hz)
1.7 (MES)
3.3 (6 Hz)
1.8 (scPTZ)
252.7
130.6
239.4
430.7
VPAg
0.5
(220.1–290.2) (117.6–145.2) (209.2–274.1) (407.9–454.9)
Values in parentheses are 95% confidence intervals determined by probit analysis.22
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 16 of 38
1
2
3
4
5
6
7
8
9
a Time to peak effect.
b ED50 (MES-maximal electroshock seizure test).
c ED50 (6 Hz-seizure test, 32 mA).
d ED50 (scPTZ- subcutaneous pentylenetetrazole seizure test).
e TD50 (NT-acute neurological deficit determined in the rotarod test).
f Protective index (TD50/ED50).
g Reference AEDs: Ethosuximide (ETX), Lacosamide (LCS), Levetiracetam (LEV), and Valproic acid (VPA)
tested in the same conditions. TPEs for model AEDs taken from own experiments or literature.23
A dash indicates - not tested, n.a. - non-active.
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
The obtained quantitative data indicated five compounds 11, 13, 19, 22, and 28 as broad-
spectrum anticonvulsants, which were effective in all the tests performed. These compounds
revealed generally higher activity in electrically evoked seizure models, such as MES and 6 Hz
(32 mA), compared to chemically induced convulsions, scPTZ test. Among the aforementioned
compounds, the most promising was compound 22, which demonstrated distinctly more potent
anticonvulsant activity in the MES, 6 Hz (32 mA), and scPTZ models than that observed for
the broad-spectrum AED, that is VPA, with ~10-fold (MES), ~6-fold (6 Hz, 32 mA), and ~4-
fold (scPTZ) higher potency. Moreover, 22 displayed substantially better therapeutic window
(PI values) than those of VPA. Notably, VPA is still recognized as one of the most relevant
first-line AEDs used for treatment of all types of seizures in patients. Compound 22 exhibited
a lower potency than that of LCS (a dual-acting AED (MES/6 Hz (32 mA)) and LEV, which
was found to be effective only in the 6 Hz (32 mA) model. Compound 22 was found to have
a wider therapeutic window (PI) in the MES test and revealed comparable safety profile in
the 6 Hz (32 mA) model than that of LCS. Unfortunately, compound 22 showed distinctly
worse PI value than that of LEV, which is well-known to have an excellent tolerability profile,
also clinically. Furthermore, compound 22 was found to be more potent in the scPTZ test than
both ETX and VPA, which are recognized as reference AEDs for this model of chemically-
induced seizures. In the scPTZ test, an anticonvulsant effect is considered when an animal is
protected from the clonic seizures with loss of righting reflex. Figure 2 shows that compound
22, compared to vehicle-treated group, prolonged the latency time to first seizure episode in a
dose-dependent manner, from 485.4 ± 57.6 s (control) to 827.8 ± 218.9 s (70.54%), 1318 ±
ACS Paragon Plus Environment

Page 17 of 38
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
253.1 s (171.53%), and 1504 ± 199.2 s (209.85%) for doses 30, 60, and 80 mg/kg, respectively.
Similar effect was observed for ETX and VPA; however, these scPTZ-active AEDs were
effective in distinctly higher doses. Efficient dose-dependent prolongation of the latency time
to first episode of clonic seizures was also observed for 11 and 19 (see Figure S1). Importantly,
the most effective compounds (11, 13, 19, 22), and reference AEDs effective in the scPTZ
model decreased the number and severity of seizures and decreased the lethality of mice in the
aforementioned test (Table S4).
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
Figure 2. The latency time to first PTZ seizure episode for 22, and reference AEDs: ETX and
VPA. Each value represents the mean ± SEM obtained from 6–10 mice. Statistical analysis:
one-way analysis of variance (ANOVA), followed by Dunnett’s post hoc test. The compounds
were administered i.p. 0.25 h (ETX) or 0.5 h before the test (22, VPA). Significant difference
compared to the control group: **p < 0.01, ***p < 0.001. Data for ETX and VPA from ref.13
The 3-Cl (13) and 3-CH3 (19) derivatives also showed potent and broad-spectrum
anticonvulsant activity. Despite relatively strong neurological impairment, these derivatives
displayed higher activity and also better PIs than that of VPA, but unfortunately showed weaker
activity specifically in the MES and/or 6 Hz (32 mA) seizure models as well as less favorable
PIs than that of LCS and LEV. Interesting results were obtained for 3-SCF3 (27) and 3,5-diCl
(44) derivatives that appeared to be especially effective in the 6 Hz (32 mA) test. In this seizure
model, 27 and 44 were equally effective as LEV, and furthermore 44 did not affect motor
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 18 of 38
1
2
3
4
5
6
7
8
9
coordination of mice up to dose of 300 mg/kg that yielded a very favorable PI > 17 value. The
other molecules tested revealed moderate efficacy (13, 28), provided more narrow protection
as demonstrated in the MES and 6 Hz seizures (25, 32, 35, and 48), or revealed activity only in
the MES test (23). Notably, the presence of 3-OCF3 group (25) provided potent protection in
the MES and 6 Hz (32 mA) seizure models, albeit with an increased neurological impairment.
In summary, the introduction of the phenyl substituent in the side position on the acetamide
linker, which was the most important structural modification proposed in this study (see Figure
1A), caused distinctly improved protection in the MES test for 22 compared to the respective
methyl analog KA-11.10 This modification also allowed to retain high activity in the 6 Hz (32
mA) and scPTZ seizure models.
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
Antinociceptive activity. AEDs are of great importance in the pharmacotherapy of
neuropathic pain and other neurological pain syndromes. Taking into consideration similarities
in pathophysiology of epilepsy and pain (in particular neuropathic one), as well as the broad-
spectrum and potent anticonvulsant activity of compound 22, we hypothesized that it may also
be a potent analgesic agent. To confirm this, we tested compound 22 for analgesic properties in
the formalin test of tonic pain in mice. VPA was used as the reference AED, which showed
broad-spectrum anticonvulsant activity, similar to compound 22, in preclinical experiments
(e.g., MES, scPTZ, and 6 Hz test models). Pregabalin (PGB) was used as the second reference
compound to compare the activity of 22. In fact, PGB (together with gabapentin) is one of the
two the most often used AEDs in the clinical management on neuropathic pain. Therefore, the
antinociceptive activity of compound 22 may provide an essential clinical advantage over VPA,
which is widely used in the pharmacotherapy of seizures, but its role in the treatment of pain is
much less significant.
The formalin test is one of the most useful screening models for testing potentially
clinically-relevant antinociceptive molecules because of its ease of administration and
ACS Paragon Plus Environment

Page 19 of 38
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
standardization and validation with reference drugs.24 Subcutaneous injection of formalin
results in a focal injury that stimulates and then damages sensory endings. Two distinct phases
of nociceptive response are associated with immediate activation of nociceptors and
sensitization of spinal reflex circuits during phase I and phase II responses, respectively.25
Moreover, it has been recently suggested that formalin injection results in pathological changes
that resemble those observed in nerve injury.26 Altered functions of sodium and calcium
channels are probably involved in the mechanism of the both phases, thus making the formalin
test a valuable method in the investigation of analgesic properties of anticonvulsant agents
affecting sodium and/or calcium currents. Sodium channel blockers preferentially attenuate the
nociceptive response in phase II of the test, whereas gabapentin or pregabalin (PGB), which
bind to alpha2-delta (α2δ) site of voltage-gated calcium channels, are active in both phases.27
As it is presented in Figure 3, the i.p. administration of compound 22 prior to the subcutaneous
injection of formalin significantly attenuated the nociceptive response in mice in both phases
of the test. Its ED50 value in phase I was found to be 28.5 mg/kg, whereas the ED50 value in
phase II was found to be 12.4 mg/kg. Compound 22 was distinctly more effective in both phases
compared to previously described KA-11 (ED50 (phase I) = 71.2 mg/kg and ED50 (phase II) =
29.3 mg/kg) (for KA-11 structure see Figure 1).11 The results obtained in our laboratory for
VPA revealed that this drug was ineffective in phase I, and the activity in phase II was moderate
(ED50 = 132.9 mg/kg)9, and much lower than that obtained for compound 22. The
administration of PGB resulted in a significant decrease of nociceptive response in the early
phase only at the highest dose of 30 mg/kg, which is in contrast with the potent activity of
compound 22. Moreover, PGB was not as active as compound 22 in the late phase of the test
(ED50 = 20.97 mg/kg). Taking into account the fact that PGB is clinically useful in the treatment
of neuropathic pain, these results suggested highly relevant analgesic properties of compound
22.
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
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 20 of 38
1
2
3
4
5
6
7
8
9
10
11
12
13
14
15
16
17
18
19
20
21
22
23
24
25
26
27
28
29
30
31
32
33
34
35
36
37
38
39
40
41
42
43
44
45
46
Figure 3. Figure 3. Antinociceptive activity of compound 22 and reference drugs; valproic acid (VPA), pregabalin (PGB) in the formalin test.
Results are shown as time of licking in phase I (0–5 min after intraplantar injection of formalin) and in phase II (15–30 min after formalin injection).
Each value represents the mean ± S.E.M. for 8-10 animals. Statistical analysis: one-way ANOVA followed by post hoc Dunnett's test. Statistical
significance compared to vehicle-treated animals (Tween): *pꢀ<ꢀ0.05, **p<0.01, ***pꢀ<ꢀ0.001, ****pꢀ<ꢀ0.0001. C-control group.
ACS Paragon Plus Environment

Page 21 of 38
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
Influence on spontaneous locomotor activity. We tested the influence of compound
22 on spontaneous locomotor activity to investigate its sedative properties. High sedative
activity of anticonvulsants is considered to be an undesirable property, which may lead to an
incorrect or ambiguous interpretation of the in vivo results. Moreover, it may prove to be an
additional limitation in case of potential clinical application of new drug candidates. The data
presented in the Figure S2 indicate that compound 22 has a moderate sedative activity (ED50 =
22.7 mg/kg), which is comparable to these dose producing analgesic effects. Therefore, the
activity of compound 22 may be at least partially associated with sedative properties, however
these effects do not dominate over its analgesic or anticonvulsant properties. Our observations
are supported by the results obtained with PGB, which also decreased spontaneous locomotor
activity of mice at the analgesic dose of 30 mg/kg. The obtained results show that the moderate
sedative properties of this drug may be acceptable at the preclinical as well as clinical level.
In vitro radioligand binding studies and functional assays. The traditional discovery
and development process of potential therapeutic agents for the treatment of epilepsy often
involves testing in animal seizure models prior to the determination of their possible mechanism
of action.28 This strategy enables to identify compounds acting on either well-known molecular
targets for anticonvulsants or compounds with novel and unique mechanism of action. In
addition, the currently used AEDs possess various mechanisms of action, and notably in many
cases, their precise pharmacodynamic profiles were identified after their regulatory approval
and introduction to the market (e.g. levetiracetam).
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
Compound 22 demonstrated broad-spectrum anticonvulsant and analgesic activity, thus,
in order to identify the potential targets of compound 22, we carried out binding studies with
several voltage-gated or ligand-gated channels and GABA-transporter as the most common
molecular targets for anticonvulsants. Moreover, considering the potential pro-arrhythmic
activity, the interaction of 22 with potassium channel (hERG) was also investigated (Table 2).
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 22 of 38
1
2
3
4
Table 2. In vitro binding/functional assays for 22 and (concentration of 100 μM)
5
6
7
8
% Inhibition of control
Binding studies
Source
specific bindinga
rat cerebral
cortex
Na+ channel (site 2)
82.5
9
L-type Ca2+
(dihydropyridine site, antagonist
radioligand)
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
rat cerebral
cortex
82.3
69.6
58.2
L-type Ca2+
(diltiazem site, antagonist
radioligand)
rat cerebral
cortex
L-type Ca2+
(verapamil site, antagonist
radioligand)
rat cerebral
cortex
rat cerebral
cortex
NMDA (antagonist radioligand)
26.2
21.0
5.2
Na+ channel (site 1)
rat brain
N-type Ca2+
(antagonist radioligand)
rat cerebral
cortex
GABA transporter
(antagonist radioligand)
rat cerebral
cortex
1.1
human recombinant
(HEK-293 cells)
Potassium channel (hERG)
25.8
% Inhibition of control
agonist responsea
Functional studies
TRPV1 (VR1) (h)
(antagonist effect)
human recombinant
(CHO cells)
71.7
Cav1.2 (L-type) (h) calcium ion
channel cell based antagonist
calcium flux assay
human recombinant
(HEK-293 cells)
104.0
a
Results showing activity higher than 50% are considered to represent significant effects of the test
compounds; results showing an inhibition between 25% and 50% are indicative of weak effect; results showing
an inhibition lower than 25% are not considered significant and mostly attributable to variability of the signal
around the control level. Binding or functional studies were performed commercially in Cerep Laboratories
(Poitiers, France). A dash indicates - not tested.
According to the obtained results, compound 22 revealed significant binding to voltage-
gated sodium channels (site 2) and calcium channels (L-type, three sites) at a high concentration
of 100 μM. Importantly, in case of the calcium channels, the highest affinity was observed for
the dihydropyridine site, then the diltiazem site, and finally the verapamil site. Furthermore, the
L-type calcium ion channel cell-based flux studies (see Table 2) revealed that 22 possesses an
ACS Paragon Plus Environment

Page 23 of 38
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
antagonist activity. It should be stressed here that L-type voltage-gated calcium channels are
widely distributed in the CNS, mostly post-synaptically and are responsible for regulating of
neuronal firing. Notably, the most recent data indicate the involvement of central L-type
calcium ion channels in pathological states associated with persistent pain (including
neuropathic one).29,30 The 22 molecule did not interact or revealed only weak interaction with
NMDA receptors, voltage-gated sodium channels (site 1), N-type calcium channels, GABA
transporter, and notably with potassium channel (hERG) at concentration of 100 μM; thus, we
can conclude that it has a low risk of pro-arrhythmic activity. It should be stressed here that the
aforementioned binding results, may only suggest voltage-gated sodium and calcium channels
(L-type) as potential molecular targets for hybrid compounds reported herein. Consequently,
there is a probability that both anticonvulsant and antinociceptive activity may result from
additional, and currently undefined mode of action.
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
Compound 22 shares some structural similarities (e.g. aromatic ring position and amine
fragment) with that of several TRPV1 antagonists (e.g., BCTC, see Figure 1). Therefore, in the
next step, we determined the antagonist activity of compound 22 against TRPV1 receptor.
Importantly, several recent neurobiological studies implicate central TRPV1 in the induction of
epileptic seizures, and they have suggested that the inhibition of aforementioned channel is a
new mechanism of action for anticonvulsants.31,32 The results of functional assays (see Table 2
and Figure S3) confirmed the TRPV1 channel antagonist activity of compound 22 (IC50=44
μM, KB= 5.8 μM). Therefore, we hypothesize that TRPV1 antagonism may be involved,
especially in the antinociceptive or/and in the anticonvulsant activity of compound 22. Notably,
it is suggested that TRPV1 desensitization, besides the influence on sodium and calcium
conductance, is one of the crucial mechanisms of action responsible for anticonvulsant activity
of cannabidiol (CBD),33,34 which was proved to be effective against the drug-resistant epilepsy
(especially in Dravet and Lennox-Gastaut syndromes in children).35,36 It is worthy of note that
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 24 of 38
1
2
3
4
5
6
7
8
9
the comparison of the binding/functional profile for active as anticonvulsants vs. inactive
compounds, 27, 44 and 26 respectively, proved distinctly more p otent interaction with calcium
channels (L-type), TRPV1 receptors as well as voltage-gated sodium channels (site 2) for 27
and 44, which showed a potent protection in the MES and 6-Hz (32 mA) seizure tests (Table
S5). In summary, the in vitro studies may indicate a multifunctional mechanism of action for
the lead compound 22 that may explain its broad pharmacological activity in vivo.
In vitro ADME-Tox assays. These studies are performed during the early stages of the
drug discovery process, which enables us to evaluate the drug-like properties of the new
compounds in parallel to the investigation of their efficacy. The selected ADME-Tox
parameters of the most promising compound, 22, were assessed in vitro using cytochrome
(CYP) enzymes and human liver microsomes (HLMs). The results obtained summarized in
Table 3.
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
Table 3. ADME-Tox parameters of 22 determined by in vitro methods
5.8a
Human Clint (mL/min/kg)
dehydrogenation (M1)b
Metabolic pathways in human
hydroxylation (M2)
hydroxylation with ketone reduction (M3)
92.7 ± 8.2c
CYP3A4 activity (% of control ± SD at 10 μM)
CYP2D6 activity (% of control ± SD at 10 μM)
120.4 ± 2.9c
a Data from four independent reactions, each terminated at different time point (5, 15, 30 or 45 min).
b Main metabolic pathway.
c Tested in triplicate.
The in vitro metabolic stability of compound 22 was determined by using HLMs. The
Clint values were calculated by measuring the disappearance of compound 22 with time,
according to the previously described protocols and formulas.37–39 The results have shown an
excellent, low Clint values (5.8 ml/min/kg) after incubating compound 22 with HLMs.
Moreover, the metabolic pathways and the most probable structure of metabolites were
ACS Paragon Plus Environment

Page 25 of 38
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
determined by using ultra performance liquid chromatography-mass spectrometer (UPLC-MS),
supported by the predicted in silico (MetaSite 5.1.1 program), which indicates the most
probable sites of metabolism of compound 22. As a result, three metabolites M1–M3 were
obtained after the process of dehydrogenation (M1), hydroxylation (M2), and hydroxylation
with ketone reduction (M3) and were identified using in vitro models (Table 3, Figures S4–S8).
We performed luminescence-based assays for the most important isoforms of
cytochrome P450, namely, CYP3A4 and CYP2D6, which are responsible for the metabolism
of ~40%−50% of all marketed drugs. As a result, the effect of compound 22 on the activity of
CYPs may be a potential source of clinically important DDIs. Both CYP3A4 and CYP2D6
assays were performed according to the manufacturer’s protocol (see Supplementary material).
According to the results, compound 22 showed no effect on CYP3A4, whereas CYP2D6
showed slight induction effect at concentrations ≥10 µM (Table 3) (Figures S9 and S10). In
conclusion, none or slight effect of compound 22 on the most relevant CYP subtypes, as well
as its influence on Pgp (P-glycoprotein) efflux transporter activity determined only at the
highest used concentration 100 µM, indicates low risk of potential DDI.
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
CONCLUSION
In the present study we have synthetized and characterized a new series of
anticonvulsants with hybrid structure based on the pyrrolidine-2,5-dione core fragment. The
preclinical data in mice revealed that these hybrid compounds exhibited potent protection and
showed broad-spectrum activity in different animal models of seizures, such as the MES,
scPTZ, and 6 Hz (32 mA). The most promising anticonvulsant properties were found for
compound 22, which revealed its potential as a broad-spectrum antiepileptic agent. Notably,
compound 22 showed better protection and more favorable safety profile than that of VPA,
which is recognized as the most clinically valuable broad-spectrum AED. The more detailed
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 26 of 38
1
2
3
4
5
6
7
8
9
pharmacological evaluation indicated a potent antinociceptive activity of compound 22 in the
formalin-induced tonic pain model in mice. The results from in vitro studies may suggest that
the mechanism of action of compound 22 is probably multifunctional. Functional assays
showed antagonist activity of the tested compound toward the TRPV1 receptor and L-type
calcium channels. Compound 22 revealed high metabolic stability on HLMs. This molecule
showed no or very slight influence on CYP3A4, CYP2D6, but only at high concentration.
Overall, the obtained results indicate that compound 22 may be a candidate for further
development with potential therapeutic utility in epilepsy and neuropathic pain conditions.
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
MATERIALS AND METHODS
Chemistry. All reagents and solvents came from commercial suppliers. Melting points
were measured in melting point apparatus (Büchi Labortechnik, Flawil, Switzerland) in open
capillaries. The purity/homogeneity of the intermediates and target compounds was assessed
by TLC and UPLC methods. TLC was performed on silica gel 60 F254 pre-coated aluminum
sheets (Macherey-Nagel, Düren, Germany). The following developing systems were used: S1–
DCM:MeOH (9:0.2; v/v), S2–DCM:MeOH (9:0.3; v/v), S3–DCM:MeOH (9:0.5; v/v), S4–
DCM:MeOH (9:1; v/v). The spots were observed in UV light (λ = 254 nm). The mass spectra
were obtained on Waters ACQUITY™ TQD system (Waters, Milford, CT, USA) equipped
with the MS-TQ detector and UV-Vis-DAD eλ detector. The ACQUITY UPLC BEH C18, 1.7
μm (2.1  100 mm) column was used with the VanGuard Acquity UPLC BEH C18, 1.7 μm
(2.1  5 mm) (Waters, Milford, CT, USA). For UPLC/MS analyses the 1 mg/mL solutions of
each substance were prepared MeCN/water mixture (1:1; v/v). The following conditions were
applied: eluent A (water + 0.1% HCOOH), eluent B (MeCN + 0.1% HCOOH), flow rate of 0.3
mL/min, gradient of 5–100% B over 10 min, injection volume - 10 μl. The UPLC retention
times (tR) are given in minutes. Preparative column chromatography was carried out using silica
ACS Paragon Plus Environment

Page 27 of 38
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
gel 60 (particle size 0.063–0.200; 70–230 Mesh ATM) (Macherey-Nagel, Düren, Germany).
Elemental analyses (C, H, and N) for target compounds were performed using the elemental
Vario EI III Elemental analyzer (Hanau, Germany). The results obtained were within ±0.4% of
the calculated values. The NMR spectra were recorded in a Varian Mercury spectrometer
(Varian Inc., Palo Alto, CA, USA), in CDCl3 or DMSO-D6 operating at 300 MHz (1H NMR),
75 MHz (13C NMR), and 282 MHz (19F NMR) or in a JEOL-500 spectrometer (JEOL USA,
Inc. MA, USA), in CDCl3 operating at 500 MHz (1H NMR) and 126 MHz (13C NMR).
Chemical shifts are given as δ values (ppm). TMS δ = 0 (1H) was used as internal standard. The
following abbreviations were applied in the spectra description: s (singlet), br. s (broad singlet),
d (doublet), dd (double doublet), ddd (double double doublet), t (triplet), dt (doublet of triplets),
td (triplet of doublets), m (multiplet).
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
Synthetic procedure for succinamic acids 1-5. Succinic anhydride (3.0 g, 30 mmol, 1
eq) was dissolved in 15 ml of glacial acetic acid and afterwards equimolar amount of DL-
phenylglycine (4.53 g) DL-phenylalanine (4.95 g), DL-(3-fluorophenyl)glycine (5.10 g), DL-
(4-fluorophenyl)glycine (5.10 g) or DL-(3,4-difluorophenyl) glycine (5.61 g) was added. The
reaction mixture was heated to 70°C and next stirred for 12 h. After this time acetic acid was
evaporated to dryness. The intermediates 1–5 were obtained in solid form after wash up with
diethyl ether.
4-((Carboxy(phenyl)methyl)amino)-4-oxobutanoic acid (1): White solid. Yield: 87% (6.55
g); mp. 199.4–200.6ºC; TLC: Rf = 0.25 (S4); UPLC (purity 100%): tR = 2.77 min. LC-MS (ESI):
m/z calcd for C12H13NO5 (M+H)+ 252.08, found 252.1.
Synthetic procedure for monocarboxylic acids 6-10. To a suspension of the
appropriate succinamic acid 1–5 (20 mmol, 1 eq) in dry benzene (100 mL) ZnCl2 (2.73 g, 20
mmol, 1 eq) was added, and the reaction mixture was heated to 80°C. Afterwards, HMDS (4.84
g, 6.25 mL, 30 mmol, 1.5 eq) in dry benzene (15 mL) was instilled over 30 min, and the reaction
ACS Paragon Plus Environment

ACS Chemical Neuroscience
Page 28 of 38
1
2
3
4
5
6
7
8
9
was continued in reflux for additional 12 h. Next, organic solvent was evaporated under reduced
pressure. After concentration, the residual solid was dissolved in DCM and extracted with 0.1
N HCl (3 × 50 mL), water (3 × 50 mL), saturated brine (3 × 50 mL). The organic solvent was
dried over anhydrous Na2SO4 and afterwards evaporated to dryness. Intermediates 6–10 were
obtained as solids after wash up with diethyl ether.
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
2-(2,5-Dioxopyrrolidin-1-yl)-2-phenylacetic acid (6): White solid. Yield: 90% (4.20 g); mp.
198.2–199.5ºC; TLC: Rf = 0.45 (S4); UPLC (purity 100%): tR = 3.41 min. LC-MS (ESI): m/z
calcd for C12H11NO4 (M+H)+ 234.08, found 234.1. 1H NMR (300 MHz, DMSO-D6) δ 2.73 (s,
4H, imide), 5.76 (s, 1H, CH-COOH), 7.26–7.35 (m, 3H, ArH) 7.36–7.45 (m, 2H, ArH), 13.22
(br. s, 1H COOH).
Synthetic procedure for monocarboxylic acids 50 and 51. Succinic anhydride (2.0 g,
20 mmol) was suspended in 5 mL of water and 3-aminobenzoic acid or 4-aminobenzoicacid
(2.74 g, 20 mmol) was gradually added. The mixture was heated at 180°C for 1 h (after distilling
of water) in a term–regulated sand bath (ST 72 Roth, Karlsruhe, Germany). The oily crude
products were purified by crystallization from MeOH.
3-(2,5-Dioxopyrrolidin-1-yl)benzoic acid (50): White solid. Yield: 78% (3.42 g); mp. 236.2–
237.1ºC; TLC: Rf = 0.71 (S4); UPLC (purity 92%): tR = 2.73 min. LC-MS (ESI): m/z calcd for
C11H9NO4 (M+H)+ 220.05, found 220.2. 1H NMR (300 MHz, DMSO-D6) δ 2.79–2.94 (m, 4H,
imide), 7.46–7.66 (m, 2H, ArH), 7.89–8.17 (m, 2H, ArH), 11.75 (s, 1H, COOH).
Synthetic procedure for target compounds 11–47, 52, and 53. Carbonyldiimidazole
(1.17 g, 7.2 mmol, 1.2 eq) was dissolved in 5 mL of dry DMF. Afterwards, this solution was
added to solution of given intermediate 6–10, 50 or 51 (6 mmol, 1 eq) in 10 mL of anhydrous
DMF (while stirring). After 0.5 h the appropriate secondary amine (6 mmol, 1 eq) in 5 mL of
anhydrous DMF was instilled. The reaction was continued for approximately 24 h at an ambient
temperature and next organic solvent was evaporated to dryness. The column chromatography
ACS Paragon Plus Environment

Page 29 of 38
ACS Chemical Neuroscience
1
2
3
4
5
6
7
8
9
was applied for purification of crude products, using the following developing systems: S1 (13,
16, 22, 31), S2 (11, 12, 14–21, 23–25, 28, 30, 33, 42–44, 52, 53), S3 (26, 27, 29, 32, 34–41, 45–
49). The desired compounds were obtained as solids after evaporation of organic solvents.
1-(2-Oxo-1-phenyl-2-(4-phenylpiperazin-1-yl)ethyl)pyrrolidine-2,5-dione (11). White
solid. Yield: 84% (1.90 g); mp. 156.7–157.4ºC; TLC: Rf = 0.35 (S2); UPLC (purity 100%): tR =
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
1
5.93 min. LC-MS (ESI): m/z calcd for C22H23N3O3 (M+H)+ 378.17, found 378.2. H NMR
(300 MHz, CDCl3) δ 2.58–2.81 (m, 5H; 4H, imide, 1H, piperazine), 2.95–3.15 (m, 2H,
piperazine), 3.17–3.42 (m, 3H, piperazine), 3.63–3.76 (m, 1H, piperazine), 3.92–4.05 (m, 1H,
piperazine), 6.12 (s, 1H, CHCO), 6.80–6.91 (m, 3H, ArH), 7.19–7.28 (m, 2H, ArH), 7.29–7.47
(m, 5H, ArH); 13C NMR (75 MHz, CDCl3) δ 28.1, 42.4, 45.8, 48.9, 49.2, 56.8, 116.5, 116.6,
120.6, 128.6, 128.6, 128.9, 129.1, 129.2, 129.8, 129.9, 133.0, 150.7, 165.0, 176.3. Anal. calcd
for C22H23N3O3 (377.44): C: 70.01, H: 6.14, N: 11.13; Found C: 70.10, H: 6.18, N: 11.10.
Synthetic procedure for starring 4-phenylpiperazine derivatives A13–A24.
Appropriate aryl bromide (10 mmol, 1 eq), Pd2dba3 (0.37 g, 0.4 mmol, 0.04 eq), BINAP (0.37
g, 0.59 mmol, 0.06 eq), sodium tert-butoxide (1.35 g, 14 mmol, 1.4 eq) and Boc-piperazine
(3.74 g, 20 mmol, 2 eq) were suspended under inert gas (nitrogen) atmosphere in 50 mL of dry
toluene. The reaction mixture was refluxed for 12 h, subsequently cooled and filtered through
Celite 545 Merck (Darmstadt, Germany). The organic layer was washed with water and next
with brine, dried over anhydrous Na2SO4, an finally concentrated in vacuo. The Boc-protectd
amines A1–A12 were purified by column chromatography using the following developing
systems: S2 (A1, A3–A4, A6–A12) or S3 (A2, A5). The aforementioned intermediates were
obtained as light oils. In the next step, A1–A12 were dissolved in 5 mL of DCM, treated with
TFA (3 eq) and stirred at room temperature for 5 h. Afterwards, the organic solvents were
evaporated in vacuo. The resulting oil residue was dissolved in water (20 mL) and then 25%
ammonium hydroxide was carefully added to pH = 8. The aqueous layer was extracted with
ACS Paragon Plus Environment