Discovery of a Potent, Selective T-type Calcium Channel Blocker as a Drug Candidate for the Treatment of Generalized Epilepsies

Article abstract of DOI:10.1021/acs.jmedchem.7b01236

We report here the discovery and pharmacological characterization of N-(1-benzyl-1H-pyrazol-3-yl)-2-phenylacetamide derivatives as potent, selective, brain-penetrating T-type calcium channel blockers. Optimization focused mainly on solubility, brain penetration, and the search for an aminopyrazole metabolite that would be negative in an Ames test. This resulted in the preparation and complete characterization of compound 66b (ACT-709478), which has been selected as a clinical candidate.

Full text of DOI:10.1021/acs.jmedchem.7b01236

Subscriber access provided by READING UNIV
Article
Discovery of a Potent, Selective T-type Calcium Channel Blocker
as a Drug Candidate for the Treatment of Generalized Epilepsies
Olivier Bezençon, Bibia Heidmann, Romain Siegrist, Simon Stamm, Sylvia Richard,
Davide Pozzi, Olivier Corminboeuf, Catherine Roch, Melanie Kessler, Eric A. Ertel,
Isabelle Reymond, Thomas Pfeifer, Ruben de Kanter, Michael Toeroek-Schafroth, Luca G.
Moccia, Jacques Mawet, Richard Moon, Markus Rey, Bruno Capeleto, and Elvire Fournier
J. Med. Chem., Just Accepted Manuscript • DOI: 10.1021/acs.jmedchem.7b01236 • Publication Date (Web): 08 Nov 2017
Downloaded from http://pubs.acs.org on November 9, 2017
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 free 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 accessible to all
readers and 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.
Journal of Medicinal Chemistry 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 64
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Discovery of a Potent, Selective Tꢀtype Calcium
Channel Blocker as a Drug Candidate for the
Treatment of Generalized Epilepsies
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
Olivier Bezençon*, Bibia Heidmann, Romain Siegrist, Simon Stamm, Sylvia Richard, Davide
Pozzi, Olivier Corminboeuf, Catherine Roch, Melanie Kessler, Eric A. Ertel, Isabelle Reymond,
†
Thomas Pfeifer , Ruben de Kanter, Michael Toeroek-Schafroth, Luca G. Moccia, Jacques
Mawet, Richard Moon, Markus Rey, Bruno Capeleto, and Elvire Fournier
Chemistry, Biology and Pharmacology & Preꢀclinical development, Drug Discovery, Idorsia
Pharmaceuticals Ltd, Hegenheimermattweg 91, CH – 4123 Allschwil
KEYWORDS: Tꢀtype calcium channels; epilepsy; Ames test; clinical candidate.
ABSTRACT: We report here the discovery and pharmacological characterization of Nꢀ(1ꢀ
benzylꢀ1Hꢀpyrazolꢀ3ꢀyl)ꢀ2ꢀphenylacetamide derivatives as potent, selective, brainꢀpenetrating Tꢀ
type calcium channel blockers. Optimization focused mainly on solubility, brain penetration, and
search for an aminopyrazole metabolite that would be negative in an Ames test. This resulted in
the preparation and complete characterization of compound 66b (ACTꢀ709478), which has been
selected as clinical candidate.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 2 of 64
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
INTRODUCTION
Tꢀtype calcium channels (TTCCs) belong to the family of the lowꢀvoltageꢀactivated calcium
channels. These channels are the products of three genes, CACNA1G, CACNA1H, and
CACNA1I , yielding the Ca 3.1, Ca 3.2, and Ca 3.3 channels, respectively. While the Ca 3.1ꢀ
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
v
v
v
v
1
and Ca 3.2ꢀchannels are expressed in several organs, in particular the heart and the brain , the
v
Ca 3.3 channel seems to be exclusively expressed in the brain. The therapeutic potential of a
v
TTCC blocker for cardiovascular diseases led to the development of weak, nonꢀselective
2
compounds, whose physicochemical properties resulted in a poor brain penetration . On the other
hand, the discovery and development of potent, selective, brain penetrant TTCC blockers has
been the result of more recent research. It should be noted that throughout the manuscript, the
term selective blockers is used for compounds that block the three TTCCs, Ca 3.1, Ca 3.2, and
v
v
Ca 3.3 with similar potencies, while being much less potent toward any other target, in particular
v
Ca 1.2.
v
The belated discovery of potent, selective TTCC blockers might be linked to difficulties in
cloning these channels, and to the absence of high throughput screening techniques for these
targets until recently. Indeed, while high throughput screening assays, like FLIPR assays, have
been known for many years in the field of Gꢀproteinꢀcoupled receptors (GPCRs) for instance, the
search for voltage gated channel blockers relied for years solely on the low throughput patchꢀ
clamp technique. With the development of automated patchꢀclamp techniques in the early 2000s,
the throughput increased slightly but it has been only recently that highꢀthroughput colorimetric
3
assays were developed to measure the activity of voltageꢀgated calcium channels.
ACS Paragon Plus Environment

Page 3 of 64
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
The discovery of TTCC blockers followed this evolution. Mibefradil (1, Fig. 1), discovered in
the early 1990s in a phenotypic manner, was considered for many years as the prototypic TTCC
blocker. On the other hand, this compound is not brainꢀpenetrant, and blocks multiple channels.
In our hands, compound 1 blocked Ca 3.1, Ca 3.2, Ca 3.3, Ca 1.2, and hERG with IC ꢀvalues
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
v
v
v
v
50
of 64 nM, 130 nM, 130 nM, 250 nM, and 580 nM, respectively. Literature data shows that this
4
compound also blocks Ca 2.2 (IC ꢀvalues between 1000 and 3000 nM) , as well as Na 1.5,
v
50
v
5
Na 1.4, Na 1.2, and Na 1.7 (IC ꢀvalues between 500 and 2000 nM). Many years later, two
v
v
v
50
6
selective, brainꢀpenetrant TTCC blockers, MKꢀ8998 (2a) and Zꢀ944 (2b), entered clinical trials.
More recently, CXꢀ8998 and CXꢀ5395 (structures unknown) were claimed to be new TTCC
7
blockers in clinical trials . A few model compounds were shown to be efficacious in animal
8
9
6a, 10
11
8f,
models for generalized epilepsy , insomnia , schizophrenia
, Parkinson , essential tremor
12
13
14
,
and obesity . Compound 2b shows efficacy in preclinical pain models . The potential
therapeutic benefit of a TTCC blocker in absence epilepsies is supported by genetic, functional
or pharmacological evidence. All TTCCs are expressed in the neuronal network involved in the
15
generation of the spikeꢀandꢀwave discharges, which is the hallmark of absence seizures .
Mutations in the genes expressing Ca 3.1 or Ca 3.2 have been described both in human and
v
v
1
6
animal models of absence epilepsy or other forms of idiopathic generalized epilepsies . Most
mutations in Ca 3.2 were described as gainꢀofꢀfunction mutations due to an increase of intrinsic
v
activity of the channels or to an increase of trafficking of the channels to the plasma membrane.
Ca 3.1 overexpression or, in contrast, knockout, was shown to modulate the generation of spikeꢀ
v
1
7
andꢀwave discharges in mice . Less is know regarding the role of Ca 3.3.
v
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 4 of 64
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 1. Structures of 1, 2a , and 2b
Different classes of TTCC blockers that were studied in our laboratories were presented
8
a, 8b
elsewhere
. In particular, we found that pyrazole 3ꢀcarboxamides are subtypeꢀselective Ca 3.1
v
18
blockers, as exemplified by the derivative 3 (Fig. 2). Along structure activity relationship
(SAR) studies, we prepared the corresponding inverse amides, starting with a focused library of
amide couplings based on commercially available 3ꢀaminopyrazoles, and we identified
compound 4 as a moderately potent TTCC blocker (IC ~ 40ꢀ800 nM) with some selectivity
5
0
versus Ca 1.2 channels (IC = 2800 nM). Compound 4, with a molecular weight of 326 g/mol, a
v
50
2
clogP of 3.2, a polar surface area of 47 Å , and a ligand efficiency of 0.44 on Ca 3.1, represented
v
an acceptable starting point in the search for TTCC blockers. Starting from this compound, we
aimed at identifying a compound that could be selected as a preclinical candidate for later
treatment of generalized epilepsies in human. In a particular, beside a standard safety profile, we
wanted to develop a compound blocking all three TTCCs with similar potencies, with an
excellent efficacy in animal models at low dose, and suitable for a once a day dosing in human.
In this article, we describe our efforts in optimizing this series of aminopyrazole amides,
ACS Paragon Plus Environment

Page 5 of 64
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
eventually leading to the identification of the clinical candidate 66b (Nꢀ(1ꢀ((5ꢀcyanopyridinꢀ2ꢀ
yl)methyl)ꢀ1Hꢀpyrazolꢀ3ꢀyl)ꢀ2ꢀ(4ꢀ(1ꢀ(trifluoromethyl)cyclopropyl)phenyl)acetamide,
ACTꢀ
709478). On our way to this optimized compound, we had to adjust polarity to maintain
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
solubility without promoting active transport. Furthermore, we had to go through an extensive
optimization of the aminopyrazole in order to identify an Ames negative aromatic amine that was
formed after hydrolysis of the amide bond in plasma.
Figure 2. A first identified aminopyrazole derivative
RESULTS AND DISCUSSION
Initial SAR studies. From compound 4, we started investigating the SAR at the phenylacetic
acid moiety. Compounds 5a – 5l (Table 1) were prepared by amide coupling between a
phenylacetic derivative, and an aminopyrazole, prepared from nitropyrazole in two steps (see
supporting information). Derivative 5a, bearing a nonꢀsubstituted phenyl group, was less potent
than compound 4, whereas a position screen with a methyl group led to an excellent gain in
potency at the paraꢀposition (compound 5b). Substituting the metaꢀposition (compound 5c) was
of secondary importance, and substituting the orthoꢀposition (compound 5d) had a minor
influence only. The nature of the paraꢀsubstituent did not seem to play a crucial role (compounds
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 6 of 64
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
5
f to 5k), although an electrodonating methoxy substituent led to some loss in potency
(
compound 5e vs. 5b). The longer electrodonating ethoxyꢀ and trifluoromethoxy groups led to
the compounds 5f and 5g, equipotent to compound 5b. An electrowithdrawing chlorine atom
compound 5h) was tolerated as well. Incorporation of an isopropylꢀ or a dimethylamino group
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
(
in the paraꢀposition (compounds 5i and 5j) led to potent blockers. Also, the moderate potency of
compound 5e was greatly improved by introducing additionally a metaꢀmethyl substituent
(compound 5k). Finally, a benzoic acid substituent as replacement for the phenylacetic acid
moiety led to completely inactive compound 5l.
Table 1. Potency of aminopyrazoles on TTCCs
O
O
N
R
N
N
N
N
N
H
H
5
a - 5k
5l
Ca 3.1
Ca 3.2
Ca 3.3
v
v
v
Compound
R
a
a
a
IC in nM
IC in nM
IC in nM
5
0
50
50
5
5
5
5
5
5
5
5
5
a
b
c
d
e
f
Hꢀ
1140
35
>10000
390
pꢀMeꢀ
mꢀMeꢀ
oꢀMeꢀ
pꢀMeOꢀ
pꢀEtOꢀ
210 (52ꢀ1100)
18 (9.0ꢀ34)
54
1040 (260ꢀ5800) 67 (26ꢀ200)
490
170
35
3400
120
610
32
210
25
g
h
i
pꢀCF
3
Oꢀ
13 (4.0ꢀ61)
41
155 (88ꢀ310)
280
15 (9.0ꢀ28)
24
pꢀClꢀ
i
pꢀ Prꢀ
5.1 (3.7ꢀ7.7)
23 (10ꢀ91)
7.6 (3.5ꢀ26)
ACS Paragon Plus Environment

Page 7 of 64
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
5
5
5
j
pꢀMe
2
Nꢀ
14 (13ꢀ17)
11 (5.0ꢀ24)
>10000
38 (37ꢀ68)
2.9 (2.8ꢀ37)
6.5
k
pꢀMeOꢀmꢀMeꢀ
57 (14ꢀ300)
l
ꢀ
>10000
>10000
a
IC values based on single measurements are indicated in italics, other IC values are the
5
0
50
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
geometric mean of at least two measurements. Where sufficient data were available for
calculation 95% confidence intervals are indicated in parenthesis. Data were obtained with a
Fluorescence Imaging Plate Reader (FLIPR ) assay, see supporting information for details.
®
We then turned our attention to the Nꢀbenzyl substituent attached to the pyrazole ring (Table
2). A position screen with a methyl group demonstrated that the orthoꢀposition should not be
substituted (compound 6a vs. 5j), while substitution was tolerated at the metaꢀposition
(compound 6b) and beneficial at the paraꢀposition (compound 6c). Focusing on the paraꢀ
position, we found that electrowithdrawing substituents (compounds 6d, 6e, 6g) as well as
electrodonating groups (compounds 6f, 6h, 6i) were tolerated. Especially, very potent
compounds were obtained with a halogen at this position. Combining substituents at the paraꢀ
and metaꢀpositions led to very potent derivatives like the compounds 6j, 6k, and 6l. In contrast,
paraꢀsubstitution with hydrogen bond donors or acceptors (compounds 6m and 6n) was not
tolerated. Also, as for the other end of the molecule, replacement of the benzyl by a phenyl was
detrimental to the potency (compound 6o).
Table 2. Potency of aminopyrazoles on TTCCs
N
N
O
O
N
R
N
N
N
H
N
N
H
6a - 6n
6o
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 8 of 64
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Compound
R
Ca 3.1
Ca 3.2
Ca 3.3
v
v
v
a
a
a
IC in nM
IC in nM
IC in nM
5
0
50
50
6a
6b
6c
6d
6e
6f
oꢀMeꢀ
mꢀMeꢀ
pꢀMeꢀ
pꢀClꢀ
250 (150ꢀ420)
8.1 (4.1ꢀ16)
7.0 (4.5ꢀ11)
1.1 (0.5ꢀ2.2)
4.2 (2.0ꢀ10)
23 (14ꢀ41)
11 (5.4ꢀ32)
6.8 (2.5ꢀ21)
2.2 (1.7ꢀ2.8)
1.5 (1.0ꢀ2.5)
3.2 (1.7ꢀ6.3)
3.1 (1.5ꢀ7.8)
>10000
190 (21ꢀ3400) 40 (19ꢀ88)
36 (6.0ꢀ340)
10 (3.2ꢀ40)
4.1 (0.3ꢀ93)
12 (3.3ꢀ93)
47 (10ꢀ570)
20 (11ꢀ44)
10 (7.2ꢀ15)
5.1 (1.6ꢀ18)
3.4 (2.3ꢀ4.0)
7.8 (0.3ꢀ690)
14 (8.1ꢀ27)
>10000
2.8 (1.1ꢀ7.2)
1.8 (1.8ꢀ3.2)
1.1 (0.3ꢀ4.8)
2.0 (0.8ꢀ6.8)
3.4 (2.7ꢀ4.2)
3.5 (2.0ꢀ6.8)
1.6 (1.2ꢀ2.1)
1.0 (0.8ꢀ1.2)
0.89 (0.7ꢀ1.3)
0.92 (0.4ꢀ2.7)
1.8 (1.3ꢀ2.7)
3300
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
pꢀFꢀ
pꢀMeOꢀ
pꢀCNꢀ
6g
6h
6i
pꢀF
pꢀF
HCOꢀ
COꢀ
2
3
6j
3,4ꢀdiFꢀ
6k
6l
3,5ꢀdiFꢀ4ꢀMeOꢀ
mꢀFꢀpꢀCNꢀ
6m
pꢀCH CONHꢀ
3
6n
pꢀ(2ꢀoxopyrrolidinꢀ1ꢀ
yl)
5700
>10000
520
6o
ꢀ
1400
6300
2300
a
IC values based on single measurements are indicated in italics, other IC values are the
5
0
50
geometric mean of at least three measurements, 95% confidence intervals are indicated in
®
parenthesis. Data were obtained with a Fluorescence Imaging Plate Reader (FLIPR ) assay, see
supporting information for detail.
Characterization of a first model compound. At this stage, we selected compound 6j for
further characterization. Compound 6j appeared to be selective for TTCCs, being a weak blocker
of Ca 1.2ꢀchannels (IC = 7100 nM), and of sodium and potassium channels in rat cortical
v
50
ACS Paragon Plus Environment

Page 9 of 64
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
neurons (< 15% and <25% block at 10 ꢀM, respectively). Aqueous solubility was low (3 mg/L at
pH 7), whereas solubility in fasted, respectively fed state simulated intestinal fluid (FaSSIF, resp.
FeSSIF) remained acceptable (9 and 31 mg/L, resp.), a common characteristic for this series of
compounds. Other ADME parameters were promising, especially with an intrinsic clearance of
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
4
1 ꢀL/min/mg protein in human microsomes and 360 ꢀL/min/mg protein in rat microsomes. In
this regard, the presence of a para substituent at the phenylacetyl moiety did not only increase
potency, it also dramatically stabilized compounds against oxidative metabolism in rat liver
microsomes. Finally, compound 6j was well permeable in vitro without being a Pgp substrate (in
.
ꢀ6
MDR1ꢀMDCK cells PappAꢀB = 62 10 cm/s; ratio BꢀA/AꢀB = 0.9). We then administered this
compound or matching vehicle in a crossꢀover design by oral gavage at the beginning of the
night active period to Wistar Albino Glaxo Rats of Rijswijk (WAG/Rij rats), a model of
1
9
generalized nonꢀconvulsive absence like epilepsy . These rats had been previously implanted
with telemetry transmitters allowing continuous recording of the spontaneous seizures by
electroencephalogram (EEG) in freely moving undisturbed animals. Over the following 12ꢀh
period, 10 mg/kg of 6j completely suppressed seizures (p < 0.001, paired tꢀtest) (Fig. 3a). Brain
concentrations of 6j were measured in normal Wistar rats 1 h after the same dose: total and free
concentrations were 1703 ± 415 nM, and 22 ± 5 nM, respectively (fu,brain = 0.013). At that time
point, free brain concentrations were 6ꢀfold to 25ꢀfold above the IC ꢀvalues measured on the
5
0
TTCCs, suggesting that the channels are well inhibited during a significant fraction of the 12ꢀh
EEG recording.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 10 of 64
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Figure 3. Impact of compounds 6j (a), 16p (b), 41d (c) or 41f (d) at 10 mg/kg po on the
cumulative duration of absence seizures in male WAG/Rij rats over the first 6 h or the first 12 h
night periods following administration. Data are expressed as mean ± SEM (n = 6 to 8 per
group); *, p < 0.05; ***, p < 0.001 compared to vehicle (10% PEG400 + 90 % MC 0.5%) treated
rats (paired tꢀtest); [B]u = free brain concentration 1 h post administration in normal Wistar rats
(
mean ± SD, n = 3). Baseline values (in min) for (a) 6 h period, 6j, 20 ± 1 vs. veh., 19 ± 1; 12 h
period, 6j, 37 ± 2 vs. veh., 35 ± 2; (b) 6 h period, 16p, 29 ± 2 vs. veh., 27 ± 8; 12 h period, 16p,
6 ± 2 vs. veh., 43 ± 6; (c) 6 h period, 41d, 17 ± 1 vs. veh., 19 ± 2; 12 h period, 41d, 32 ± 3 vs.
4
veh., 34 ± 3; (d) 6 h period, 41f, 20 ± 2 vs. veh., 21 ± 1; 12 h period, 41f, 36 ± 4 v.s veh., 38 ± 3.
Tackling solubility. Overall, we had an excellent starting point for further optimization. On the
other hand, an improved solubility profile could facilitate the development of this class of
ACS Paragon Plus Environment

Page 11 of 64
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
compounds. To tackle this issue, we employed different strategies, such as decreasing the
2
0
number of aromatic systems , replacing phenyl rings with heteroaryls, or introducing polar
substituents. In terms of potency, replacing the Nꢀbenzyl substituent by heterocyclic or acyclic
systems was tolerated. Compound 9 (Scheme 1), for instance, was prepared via activation of
commercially available 2ꢀ(2,2,2ꢀtrifluoroethoxy)ethanꢀ1ꢀol and alkylation of 3ꢀnitropyrazole (→
compound 7), followed by reduction (→ compound 8) and amide coupling. The synthesis of
tetrahydropyrans 15a and 15b was more complex. Activation of commercially available (3,4ꢀ
dihydroꢀ2Hꢀpyranꢀ2ꢀyl)methanol, and Nꢀalkylation led to compound 10. Hydroboration (ꢀ
compound 11), Dess-Martin oxidation (ꢀ compound 12), and fluorination led to racemic
compound 13 with two fluoroꢀsubstituents mimicking the 3,4ꢀdifluoro pattern of compound 6j.
Both enantiomers were separated by chiral HPLC. We did not determine their absolute
configuration, and classified them arbitrarily as first and second eluting enantiomers. Subsequent
reduction to compounds 14a and 14b, respectively, and amide couplings led to the desired
products.
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 9, 15a, and 15b showed an improved solubility profile, with values in the range of
10 ꢀ 20 mg/L in an aqueous buffer at pH 7 (Table 3). However, compound 9 (as many other
compounds with an acyclic substituent on the pyrazole) was less potent than our best analogues.
Nevertheless, compounds 15a and 15b displayed potencies deserving further attention. It should
be noted that both enantiomers were equipotent, demonstrating no differentiation with regard to
stereochemistry. Unfortunately, aminopyrazole 14b, a potential metabolite of compound 15b,
was found to be positive in an Ames test, which led us to discontinue this subseries (vide infra).
Also, compound 15a, while showing reasonable stability in human liver microsomes (CL = 43
int
ꢀ
L/min/mg protein), was metabolically unstable in rat (rat liver microsomes: CLint > 500
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 12 of 64
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
ꢀ
L/min/mg protein; in vivo at 1 mg/kg iv: Cl > 100 mL/min/kg), rendering further evaluations of
this compound difficult.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 1. (i) MsCl, Et N, CH Cl , 0 °C, 1 h; (ii) 3ꢀNitropyrazole, NaH, DMF, rt, overnight; (iii)
Zn, NH Cl, acetone, water, rt, 1 h; (iv) 4ꢀIsopropylphenylacetic acid, EDC HCl, DMAP, DMF,
rt, overnight, 23%; (v) MsCl, Et N, CH Cl , 0 °C, 45 min; (vi) 3ꢀNitropyrazole, NaH, DMF, rt to
3
2
2
.
4
3
2
2
.
120 °C, 4 h, 76%; (vii) BH THF, THF, 0 °C to rt, 4 h, then 30% H O /H O, 5M aq. NaOH, 0 °C
3 2 2 2
to rt, 2 h, 30%; (viii) DessꢀMartin periodinane, NaHCO , CH Cl , rt, overnight, 15% over 2
steps; (ix) Et N 3HF, XtalFluorꢀE , CH Cl , rt, 2 h, 93%; (x) Prep. chiral HPLC (column
ChiralCel ASꢀH, heptane:EtOH 1:1); (xi) H , Pd/C, EtOAc, rt, 3 h, quant.; (xii) 4ꢀ
Isopropylphenylacetic acid, EDC HCl, DMAP, DMF, rt, 3 days, 78ꢀ86%.
3
2
2
.
®
3
2
2
2
.
Table 3. Potency of aminopyrazoles on Tꢀtype calcium channels
ACS Paragon Plus Environment

Page 13 of 64
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Compound
Ca 3.1
Ca 3.2
Ca 3.3
Sol. @ pH7
v
v
v
a
a
a
IC in nM
IC in nM
IC in nM
in mg/L
5
0
50
50
9
14 (6.8ꢀ38)
5.6 (1.6ꢀ25)
2.2 (0.9ꢀ5.7)
62 (28ꢀ180)
14 (5.9ꢀ38)
5.5 (1.7ꢀ21)
3.4 (2.5ꢀ4.7)
1.6 (0.4ꢀ8.6)
1.2 (0.6ꢀ2.3)
19
8
15a
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
15b
22
a
IC values are the geometric mean of at least three measurements, 95% confidence intervals
50
are indicated in parenthesis. Data were obtained with a Fluorescence Imaging Plate Reader
®
(
FLIPR ) assay, see supporting information for detail.
Metabolic stability, solubility, and efflux: optimizing the para-position of the phenylacetic
acid moiety. At this stage, all compounds containing an Nꢀbenzyl group, and displaying
moderateꢀtoꢀhigh solubility in an aqueous buffer at pH7 bore a dimethylaniline moiety at the
paraꢀposition of the phenylacetyl moiety. Metabolic studies on compound 5j had demonstrated
that Nꢀdemethylation was a major metabolic pathway in human hepatocytes. Since aminophenyl
derivatives are known to be potentially mutagenic, the replacement of this dimethylaniline
moiety was desirable.
For this reason, we investigated the SAR at the paraꢀposition of the phenylacetyl moiety
thoroughly. In a first step, we maintained the 3ꢀaminoꢀ1ꢀ(3,4ꢀdifluorobenzyl)pyrazole moiety
constant. Most compounds were prepared via a final amide coupling with the corresponding
carboxylic acid. The carboxylic acids themselves, if not commercially available, were prepared
in two steps from the corresponding bromophenyl or bromoheteroaryl and 2ꢀtertꢀbutoxyꢀ2ꢀ
oxoethylzinc chloride via a Negishi coupling (→ compounds of type 17, Scheme 2), followed by
hydrolysis of the tertꢀbutyl ester (→ compounds of type 18). In certain cases, a few additional
steps were necessary to prepare the bromo(hetero)aryl derivative. Thus, acid 18b was prepared
from pꢀbromostyrene via a [2+2]ꢀcycloaddition (compound 19) followed by difluorination,
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 14 of 64
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
which yielded compound 20, which in turn was sequentially transformed into ester 17b and
carboxylic acid 18b. Aminopyridines 21a and 21b were obtained by S Ar from 2,5ꢀ
N
dibromopyridine. Acid 18f was prepared in one step from the corresponding, commercially
available chloropyridine derivative. From the same starting material, an amide coupling led to
product 22, on which the cyclopropyl substituent was introduced leading to the final amide 16g.
A rhodiumꢀcatalyzed 1,4ꢀaddition yielded the ester 23, which was reduced to the aldehyde 24.
After decarbonylation, compound 25 was obtained, which was proceeded further through the
sequence reported above. Ether 26 was prepared via a Mitsunobu coupling. Ethers 27a, 27b, and
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
27c were synthesized by nucleophilic substitution of 4ꢀbromophenol or 4ꢀbromoꢀ2ꢀethylphenol,
respectively, with the corresponding mesylates or triflates.
ACS Paragon Plus Environment

Page 15 of 64
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
X
R
X
R
X
R
O
O
(i)
(ii)
HO
18b-e, 18h-i, 18k-v
O
Br
2
0, 21a-b, 25, 26, 27a-c,
or commercially available
starting material
17b-e, 17h-i, 17k-v
F
O
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
F
(iii)
(iv)
Br
Br
Br
1
9
20
F
F
N
Br
N
Br
F
(
v)
(vi)
N
N
N
N
Br
Br
Br
Br
N
21b
21a
H
(
vii)
N
N
HO
HO
(viii)
O
O
O
N
Cl
N
O
CF3
F
N
Cl
2
2
18f
(
ix)
F
B(OH)2
O
+
H
Br
O
N
N
O
(
x)
N
O
N
F
16g
O
F
O
Br
O
(xiii)
(xiv)
O
OH
OH
F
F
23
xi)
Br
(
Br
Br
26
O
O
O
Br
Br
O
27a
24
(
xii)
(xv)
O
OH
O
Br
Br
2
7b
O
(
xvi)
O
CF3
OH
Br
2
5
Br
Br
2
7c
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 16 of 64
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
t
Scheme 2. (i) ClZnCH CO Bu, Pdꢀcat., exact conditions see suppl. information; (ii)
2
2
HCl/dioxane or HCOOH, exact conditions see suppl. information; (iii) DMA, Tf O, 2,4,6ꢀ
collidine, CH ClCH Cl, 130 – 80 °C, overnight, 42%; (iv) BF Et O, DeoxoꢀFluor , CH Cl ,
toluene, 55ꢀ65 °C, 36 h, 30%; (v) (S)ꢀ3ꢀFluoropyrrolidine, DBU, DMSO, 90 °C, 7 days, 74%;
vi) 3,3ꢀDifluoropyrrolidine, DBU, DMSO, 80 °C, 1 week, 75%; (vii) NaH, trifluoroethanol, 160
C, 7 h, quant.; (viii) 1ꢀ(3,4ꢀDifluorobenzyl)ꢀ1Hꢀpyrazolꢀ3ꢀamine, HATU, DIPEA, DMF, rt,
overnight, 68%; (ix) Cyclopropylboronic acid, PEPPSIꢀIPr, K PO , toluene, 100 °C, overnight,
2
.
®
2
2
3
2
2
2
(
°
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
3
4
1
1%; (x) [Rh(COD)Cl] , KOH, dioxane, rt, overnight, 78%; (xi) DIBALꢀH, CH Cl , ꢀ78 °C, 45
2
2
2
min, quant.; (xii) [Rh(COD)Cl] , DPPP, xylene, 140 °C, 6 h, 56%; (xiii) 3,3ꢀ
Difluorocyclobutanol, DEAD, PPh , toluene, 100 °C, overnight, 92%; (xiv) (a) 3ꢀMethylꢀ3ꢀ
oxetanemethanol, MsCl, pyridine, CH Cl , rt, 4 h, 54% (b) KI, K CO , DMF, 130 °C, 2.5 h,
quant.; (xv) (a) Oxetanꢀ3ꢀol, MsCl, Et N, CH Cl , 0 °C, 4.5 h, quant. (b) KO Bu, Bu NI, DMF,
2
3
2
2
2
3
t
3
2
2
4
reflux, 18 h, 53%; (xvi) 2,2,2ꢀTrifluoroethyl trifluoromethylsulfonate, Cs CO , DMF, rt, 3 h.
2
3
Apolar substituents (Table 4, compounds 16a and 16b) led to very potent triple blockers. Not
surprisingly, such compounds were only little soluble in an aqueous buffer at pH 7. Replacement
of the phenyl group by a pyridinyl led to aminopyridinyl 16c – 16e, which were well tolerated in
terms of potency. Compounds 16c, and 16d were soluble in aqueous buffer, but moderately
potent on the Ca 3.2ꢀchannel; difluoropyrrolidine 16e increased potency on the Ca 3.2ꢀchannel,
v
v
and led to a simultaneous decrease in solubility at pH7. When combined with the pyridinyl ring,
alkoxy substituents (16f) or cycloalkyl substituents (16g) were less potent. Moving the nitrogen
atom of the pyridinyl ring to other positions, as well as the replacement of this heteroaryl with a
pyrimidinyl or a fiveꢀmembered heteroaryl were not tolerated (data not shown). Unfortunately,
these aminopyridinyl derivatives tended to be Pgp substrates, as this was the case for compound
.
ꢀ6
.
ꢀ6
1
6d (PappAꢀB = 5.1 10 cm/s, P
= 61 10 cm/s, ratio B→A/A→B = 12). The introduction of
A
ꢀ
appB
a lactame (16h) or of an oxazolidinone (16i) was well tolerated, with a marked increase in
solubility for the former example. Unfortunately, here again both compounds were Pgp
.
ꢀ6
.
ꢀ6
substrates (16h: PappAꢀB = 3.5 10 cm/s, P
= 76 10 cm/s, ratio B→A/A→B = 22; 16i:
A
appB
ꢀ
.
ꢀ6
.
ꢀ6
PappA = 11 10 cm/s, P
= 59 10 cm/s, ratio B→A/A→B = 5.5). The introduction of a
appB A
ꢀ
B
ꢀ
fused heteroaryl moiety (compounds 16j – 16n) was initially not a preferred strategy, since this
ACS Paragon Plus Environment

Page 17 of 64
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
increased the global aromaticity of the compounds. Nevertheless, indoles 16j and 16k were well
tolerated, while the solubilizing effect in aqueous buffer remained somewhat unpredictable. Both
compounds penetrated well into the brain when administered to Wistar rats at a dose of 10 mg/kg
po (sacrificed at 1 h, 16j: Ctotal,brain = 4111 ± 223 nM, CCSF = 50 ± 11 nM; 16k: Ctotal,brain = 4684
± 1701 nM, Cu,brain = 52 ± 19 nM, CCSF = 164 nM, single value). Unfortunately, these indoles
were chemically unstable, and in their amorphous form degraded within a few days at room
temperature. Introducing substituents at positions 1 and 3 led to potent compounds, but chemical
stability was not improved (data not shown). The introduction of an indazole (compound 16l) led
to a potent and soluble compound with promising brain penetration (Wistar rats, 10 mg/kg po,
sacrificed at 1 h, Cu,plasma = 72 ± 11 nM, Cu,brain = 51 ± 7 nM, CCSF = 92 ± 42 nM). Unfortunately,
a potential metabolite of this compound proved to be positive in an Ames test, such that further
investigations on compound 16d were put on hold (vide infra). Benzimidazole (compound 16m)
or a pyrrolopyridine derivative (compound 16n) were found less potent. Since we knew that
ethers were generally tolerated (compounds 5e – 5g), the SAR was extended to oxetanes (16o,
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
16p), dioxolane (compound 16q), and to more complex combination of ethers and oxetanes
(compounds 16rꢀ16t). In general, potency was maintained, and solubility increased at pH7, with
the exception of the poorly soluble compound 16r. Compound 16p was administered to Wistar
rats at 10 mg/kg po and showed good exposure in the brain after 1 h (Ctotal,brain = 2988 ± 402 nM,
Cu,brain = 63 ± 8.4 nM, CCSF = 135 ± 25 nM). In WAG/Rij rats, oral administration of 10 mg/kg
16p suppressed completely all seizures for the first 6 h following administration and when
looking over the 12 h following administration, it decreased the cumulative duration of seizures
by 78% (Fig. 3b).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 18 of 64
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
Table 4. Potency of aminopyrazoles on TTCCs
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Compound R
X
Ca 3.1
Ca 3.2
Ca 3.3
Sol.
pH 7
@
v
v
v
a
a
a
IC in nM
IC in nM
IC in nM
5
0
50
50
b
in mg/L
i
1
6a
6b
Prꢀ
CH 1.4 (1.1ꢀ1.7) 6.0 (1.6ꢀ29)
CH 1.0 (0.5ꢀ2.5) 6.1 (5.5ꢀ6.7)
1.8 (0.9ꢀ3.9)
1.8 (0.1ꢀ63)
<1
ꢀ
1
1
6c
Me
2
Nꢀ
N
N
12 (1.0ꢀ270) 46 (22ꢀ120)
2.5 (1.2ꢀ6.2) 21 (11ꢀ47)
4.0 (2.1ꢀ8.6)
5.4 (2.2ꢀ18)
25
19
16d
16e
N
1.1 (0.9ꢀ1.4) 6.7 (4.2ꢀ11)
2.3 (1.2ꢀ4.9)
3
1
6f
CF
3
CH
2
Oꢀ
N
N
10 (8.5ꢀ12) 140 (16ꢀ2200) 11 (4.2ꢀ30)
15 (9.5, 22) 220 (17ꢀ6200) 10 (5.6ꢀ20)
ꢀ
1
6g
31
1
6h
6i
ꢀ
1.3 (0.4ꢀ5.7) 3.8 (0.3ꢀ120)
1.8 (0.7ꢀ6.2)
50
6
1
ꢀ
3.4 (1.7ꢀ7.9) 15 (1.2ꢀ1700)
2.8 (1.3ꢀ7.1)
16j
ꢀ
ꢀ
1.9 (1.0ꢀ3.6) 5.3 (0.5ꢀ130)
2.3 (1.4ꢀ4.6) 14 (10ꢀ20)
1.9 (0.8ꢀ5.0)
2.1 (1.6ꢀ2.6)
2
16k
26
ACS Paragon Plus Environment

Page 19 of 64
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
1
6l
ꢀ
3.4 (1.4ꢀ4.6) 8.4 (2.8ꢀ30)
0.76 (0.4ꢀ1.8) 33
16m
ꢀ
ꢀ
45 (24ꢀ91)
7.1
120
35
9.0 (6.4ꢀ13)
3.3
ꢀ
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
16n
14
1
6o
6p
CH 2.4 (1.1ꢀ5.3) 4.3 (0.5ꢀ64)
CH 4.3 (2.3ꢀ9.6) 13 (5.0ꢀ45)
1.2 (0.7ꢀ2.5)
1.3 (0.8ꢀ2.1)
7
1
25
16q
CH 3.4 (1.8ꢀ6.8) 10 (6.0ꢀ15)
1.8 (1.2ꢀ2.8)
19
16r
CH 1.4 (0.5ꢀ4.4) 11 (1.9ꢀ110)
CH 3.5 (1.6ꢀ8.2) 34 (16ꢀ76)
CH 4.6 (0.7ꢀ54) 32 (18ꢀ57)
6.1 (2.3ꢀ18)
6.6 (4.2ꢀ10)
3.6 (2.0ꢀ6.9)
<1
16
5
16s
16t
a
IC values are the geometric mean of at least two measurements. Where sufficient data was
available for calculation 95% confidence intervals are indicated in parenthesis. Data were
obtained with a Fluorescence Imaging Plate Reader (FLIPR ) assay, see supporting information
5
0
®
for detail.
b
Single measurement
An amide bond hydrolysis leading to an Ames positive metabolite – reoptimization of the
N-benzyl moiety. While compounds such as 16l or 16p seemed suitable for further profiling, we
found that the amide bond of this class of compounds was often slowly hydrolyzed in human and
rat plasma. Compound 16k, for instance, showed a halfꢀlife time of 2 h and 5 h in human and rat
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 20 of 64
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
plasma, respectively, with the release of 3ꢀaminoꢀNꢀbenzylpyrazole 57a (see Table 9). This
metabolite was positive in the Ames test, using a bacterial reverse mutation (vide infra). A
positive Ames test does not represent a proof of mutagenicity per se, but such a result would
require extensive in vitro and in vivo studies to definitely assess the mutagenic potential of such
a compound. To facilitate further development it was thus decided that the release of an Ames
positive aminopyrazole moiety should be preferably prevented. We designed different strategies
in order to circumvent this issue: 1. Substitute the αꢀposition of the phenylacetyl moiety to
prevent hydrolysis; 2. Develop amide isosteres; 3. Introduce a methylene between the amide and
the heteroaryl; 4. Inverse the amide bond; 5. Replace the aminopyrazole moiety by another
aminoheteroaryl (pyrazole or other) that would be negative in an Ames test. In the following
paragraphs, we present the results of these different strategies.
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Substituted
α-position. Compounds 28a, and 28b were prepared via amide coupling from
commercially available starting material (see supplemental information for details). In general, a
short substituent at the αꢀposition was well tolerated (compound 28a, Table 5, compare with
compound 5i). Unfortunately, this did not stabilize the amide bond against hydrolysis (Table 5).
A double substitution generally led to more stable, but much less potent compounds (see
compound 28b). In efforts going into this direction, replacement of the amide bond by a
carbamate or a urea led to compounds that were significantly less potent as well (data not
shown).
Table 5. Potency of aminopyrazoles on TTCCs
ACS Paragon Plus Environment

Page 21 of 64
Journal of Medicinal Chemistry
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
Compound
Ca 3.1
Ca 3.2
Ca 3.3
Percentage hydrolyzed after
v
v
v
b
6
h in human plasma
a
a
a
IC in nM
IC in nM
IC in nM
5
0
50
50
5
2
2
i
5.1 (3.7ꢀ7.7)
6.2 (3.1ꢀ13)
920
23 (10ꢀ91)
7.6 (3.5ꢀ26)
4.7 (2.5ꢀ9.3)
1700
4%
8a
44 (14ꢀ160)
7%
8b
2100
<1%
a
GIC values based on single measurements are indicated in italics, other IC values are the
5
0
50
geometric mean of at least three measurements, 95% confidence intervals are indicated in
®
parenthesis. Data were obtained with a Fluorescence Imaging Plate Reader (FLIPR ) assay, see
supporting information for detail.
b
Single measurement
Amide isosteres. Replacing the carboxamide with an isostere is an obvious way to prevent any
release of an aminoheteroaryl metabolite. Therefore, we synthesized fiveꢀmembered heteroaryls
as amide isosteres, a few of which are exemplified hereafter (Scheme 3). A fused system like a
21
pyrazoloimidazole was built mainly following known literature procedures , leading to
compound 32a. Fiveꢀmembered heteroaryl, either preserving a benzylic position, like triazole
3
2b, or with the terminal phenyl ring directly linked to the isosteric ring, like oxadiazole 32c,
were prepared as well, using standard procedures via intermediates 33, 34, and 35.
Unfortunately, as depicted in Table 6, all these analogues led to substantial losses in biological
activity. While the fused system 32a maintained potencies about tenꢀfold lower than its amide
analogue 6e, compound 32b was in the high nanomolar range, and compound 32c was
completely inactive.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 22 of 64
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 3. (i) K CO , 4ꢀfluorobenzyl bromide, CH CN, rt, overnight, 69%; (ii) Aq. 2.5M
2
3
3
t
NaOH, MeOH, reflux, 1 h, 93%; (iii) DPPA, Et N, BuOH reflux, 1 h, then TFA, CH Cl , rt, 2
3
2
2
days, 87%; (iv) H , Pd/C, HCl, H O, MeOH, rt, 3 h, 37%; (v) 4ꢀ(Dimethylamino)phenylacetic
2
2
acid, HATU, DIPEA, DMF, rt, 1 h; (vi) POCl , reflux, 7.5 h, 12% over two steps; (vii)
3
.
H NNH H O, 90 °C, 2 h, 88%; (viii) 3,4ꢀDifluorobenzyl bromide, Cs CO , DMF, rt, overnight,
2
2
2
2
3
.
7
5%; (ix) K CO , nꢀBuOH, reflux, 2 days, 1%; (x) H NOH HCl, EtONa, EtOH, reflux,
2 3 2
overnight, 63%; (xi) 4ꢀIsopropylbenzoyl chloride, pyridine, toluene, reflux, overnight, 23%.
Table 6. Potencies on TTCCs
ACS Paragon Plus Environment

Page 23 of 64
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Compound
Ca 3.1
Ca 3.2
Ca 3.3
v
v
v
a
a
a
IC in nM
IC in nM
IC in nM
5
0
50
50
3
3
3
3
3
2a
2b
2c
9a
60 (20ꢀ240)
180
160 (68ꢀ420)
260
10 (6.0ꢀ16)
420
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
>10000
>10000
>10000
55 (21ꢀ160)
22 (3.8ꢀ200)
160 (37ꢀ1000)
140 (32ꢀ810)
8.0 (5.4ꢀ12)
9.8 (4.7ꢀ21)
9b
a
IC values based on single measurements are indicated in italics, other IC values are the
50
50
geometric mean of at least three measurements, 95% confidence intervals are indicated in
parenthesis. Data were obtained with a Fluorescence Imaging Plate Reader (FLIPR ) assay, see
supporting information for detail.
®
3-Aminomethylpyrazoles. We prepared 3ꢀ(aminomethyl)pyrazoles and 3ꢀ(1ꢀaminoethyl)ꢀ1Hꢀ
pyrazoles (Scheme 4). A double benzylation using commercially available starting materials led
to ester 36, which was reduced to alcohol 37. Introduction of the amine via phthalimide 38, and
subsequent amide coupling led to compound 39a. Aminoethyl derivative 39b was prepared via
2
2
aldehyde 40, which was then subjected to Ellman’s sulfinamide procedure . Unfortunately,
these derivatives were at least ten times less potent on the Ca 3.2 channel (Table 6) than the
v
corresponding 3ꢀaminopyrazole analogue (compound 16a). Despite the extensive SAR studies,
including the preparation of benzamide analogues, we were not able to obtain potent TTCC
blockers in this subseries.
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 24 of 64
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 4. (i) 3,4ꢀDifluorobenzyl bromide, Cs CO , DMF, rt, overnight, 52%; (ii) LiAlH , THF,
2
3
4
0
°C to rt, 2 h, 78%; (iii) PBr , toluene, reflux, 15 min; (iv) Cs CO , phthalimide, DMF, rt,
3 2 3
.
overnight, 75% over two steps; (v) H NNH H O, EtOH, rt, 3 h; (vi) 2ꢀ(4ꢀIsopropylphenyl)acetic
acid, EDC HCl, DMAP, DMF, rt, overnight, 88% over two steps; (vii) 3,4ꢀDifluorobenzyl
bromide Cs O , DMF, rt, overnight, 67%; (viii) 2ꢀMethylꢀ2ꢀpropanesulfinamide, Ti(OEt) , THF,
2
2
2
.
2
3
4
rt, overnight; (ix) MeMgBr, THF, 0 °C to rt, 17 h; (x) HCl, dioxane, MeOH; (xi) 4ꢀ
Isopropylphenylacetic acid, TBTU, DIPEA, DMF, rt, 30 min, 50% over 4 steps.
Inverse amides. An inverse amide would, in case of hydrolysis, lead to a pyrazole carboxylic
acid, and to a benzylic amine, which are both not expected to be Ames positive. To give a few
examples from the numerous compounds we prepared following this strategy, compounds 41a –
41d were prepared as described in Scheme 5. Carboxylic acid 43 was prepared by Nꢀalkylation
of methyl 1Hꢀpyrazoleꢀ3ꢀcarboxylate (compound 42) and subsequent saponification. Standard
amide coupling procedures with commercially available amines yielded products 41a and 41b.
Alkylation of oxetanꢀ3ꢀone led to alcohol derivative 44, which was fluorinated to compound 45.
Deprotection led to amine 46, which was coupled to acid 43 to yield amide 41c. Eventually, a
ACS Paragon Plus Environment

Page 25 of 64
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
SNAr led to nitrile 47, and subsequent reduction to amine 48. An amide coupling with acid 43
delivered product 41d.
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

Journal of Medicinal Chemistry
Page 26 of 64
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
ACS Paragon Plus Environment

Page 27 of 64
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
Scheme 5. (i) 3,4ꢀDifluorobenzyl bromide, Cs CO , DMF, 60 °C, 1 h, 52%; (ii) NaOH,
2
3
MeOH, H O, rt, 3 h, 98%; (iii) (a) (COCl) , toluene, rt, 1 h; (b) (4ꢀ
2
2
Isopropylphenyl)methanamine, CH Cl , rt, overnight, 37%; (iv) MeMgCl, nꢀBuLi, 3ꢀ
2
2
oxetanone, THF, ꢀ78 °C to rt, 6 h, 46%; (v) DAST, CH Cl , ꢀ78 °C to rt, 2 h, 49%; (vi)
2
2
HCOOH, rt, 2 h, 93%; (vii) HATU, DIPEA, DMF, rt, 1 h, 16%; (viii) 3,3ꢀDifluoropyrrolidine
hydrochloride, DIPEA, DMF, 85 °C, 22 h, 97%; (ix) RaꢀNi, H , 50 bar, NH , MeOH, 60 °C, 5
2
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
.
h, quant.; (x) DMAP, EDC HCl, DMF, rt, 6 h, 67%.
Other heterocyclic systems replacing the pyrazole moiety were prepared as well (Scheme 6).
Ethyl 2Hꢀ1,2,3ꢀtriazoleꢀ4ꢀcarboxylate was benzylated to derivative 49 (regioselectivity 2:1 in
favor of compound 49), which was saponified to carboxylic acid 50. A Negishi coupling from
ethyl 2ꢀbromooxazoleꢀ4ꢀcarboxylate and pꢀtrifluoromethoxybenzyl bromide led to oxazole 51,
which was saponified to its corresponding acid 52. A similar strategy was applied to obtain the
thiazole 54, via intermediate 53, and to prepare thiazole 56, from intermediate 55. Carboxylic
acids 50, 52, 54, and 56, were subjected to amide couplings to yield the final products 41e, 41f,
41g, and 41h respectively (Table 7).
ACS Paragon Plus Environment

Journal of Medicinal Chemistry
Page 28 of 64
1
2
3
4
5
6
7
8
9
1
1
1
1
1
1
1
1
1
1
2
2
2
2
2
2
2
2
2
2
3
3
3
3
3
3
3
3
3
3
4
4
4
4
4
4
4
4
4
4
5
5
5
5
5
5
5
5
5
5
6
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
1
2
3
4
5
6
7
8
9
0
Scheme 6. (i) 3,4ꢀDifluorobenzyl bromide, K CO , DMF, rt, 1 h, 29%; (ii) Aq. 2.5M
2
3
NaOH, EtOH, rt, 1 h, 80%; (iii) 4ꢀ(Trifluoromethoxy)benzyl bromide, activated Zn,
t
Pd( Bu P) , THF, rt, 5 h; (iv) Aq. 2M NaOH, DMF, rt, overnight, 20% over two steps; (v)
3
2
3
,4ꢀDifluorobenzyl bromide, activated Zn, Pd(PPh ) , THF, rt, overnight; (vi) Aq. 2M
3 4
t
NaOH, EtOH, THF, rt, 2 h; (vii) 3,4ꢀDifluorobenzyl bromide, activated Zn, Pd( Bu P) ,
3
2
THF, rt, 2 h, 23%; (viii) Aq. 2M NaOH, DMF, rt, 18 h, 76%.
Pyrazoles 41a, 41b, 41c, and 41d are the exact analogues to pyrazoles 16a, 16k, 16p, and 16e,
respectively. These compounds did not achieve the same potencies as their corresponding
analogues (Table 7) and, despite extensive optimization efforts, we were never able to reach this
ACS Paragon Plus Environment

Page 29 of 64
Journal of Medicinal Chemistry
1
2
3
4
5
6
7
8
9
goal. Investigation of other fiveꢀmembered heteroaryls (compounds 41e – 41h) did not yield very
potent compounds either. Nevertheless, compounds 41d and 41f were studied further (Table 8).
Metabolic stability was in an acceptable range, and aqueous solubility and unbound fraction
remained low. This translated into low levels of unbound brainꢀ and plasma concentrations, in
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
.
ꢀ6
spite of good permeability and no active transport by the Pgp pump (41d: PappBꢀA = 53 10 cm/s,
.
ꢀ6
ratio B→A/A→B = 2.8; 41f: PappB = 15 10 cm/s, ratio B→A/A→B = 1.3). In parallel with
A
ꢀ
those data, over the 12 h following oral administration in WAG/Rij rats, 10 mg/kg of 41d or 41f
did not decrease the cumulative duration of seizure (Figure 3c and d) to a similar extend as
compound 6j and 16p. Over the 6 h following administration, modest but significant decreases of
3
2% and 31% were observed for 41d and 41f respectively (p < 0.05 compared to vehicle treated
rats for both compounds, paired tꢀtest). Besides these disappointing results, we confirmed that
carboxylic acid 52 was negative in an Ames test.
Table 7. Potency of aminopyrazoles on TTCCs
F
41e
F
N
OCF3
N
N
41f
N
N
N
H
O
N
H
N
N
N
N
O
O
F
F
F
41h
F
N
41g
S
H
N
S
H
N
N
N
O
O
Compound
Ca 3.1
Ca 3.2
Ca 3.3
v
v
v
a
a
a
IC in nM
IC in nM
IC in nM
50
50
50
41a
41b
24 (8.0ꢀ84)
70 (26ꢀ210)
74 (11ꢀ760)
40
330 (180ꢀ760)
22 (5.4ꢀ120)
ACS Paragon Plus Environment